1I

Til, \, hUlll 'leu UUI (tl

/1

o AL AP o. IS
ETHNIC PROCESSES AND INTERGROUP RELATIONS IN CONTEMPORARY AFGHANISTAN

Papers

presented at the

Eleventh Annual Meeting

of the

Middle East Studies Association

at New York City

November 10, 1977

organized and edited by

Jon W. Anderson and Richard F. Strand

Occasional Paper No. 15 of the Afghanistan Council of the Asia Society

New York, Summer 1978

 TABLE OF CONTENTS

Page
Preface i

Introduction and Overview 1

Jon W. Anderson

Ethnic Competition and Tribal Schism in Eastern Nuristan 9

Richard F. Strand

Ethnic Relations and Access to Resources in Northeast 15

Badakhshan
M. Nazif Shahrani

The Impact of Pashtun Immigration on Nomadic Pastoralism in 26

Northeastern Afghanistan
Thomas J. Barfield

Religious Myth as Ethnic Boundary 35

Robert L. Canfield

References cited 43

- --
 Preface

The papers in this volume draw on recent ethnographic research in rural

Afghanistan for case material in which to examine dynamics of ethnic
identities and intergroup relations there. They are united by a focus on
the evolving interfaces between local minorities and their practical in­
corporation in a multi-ethnic state. It is our intention both to expand
the empirical base for such discussion and to help ground that discussion
in specific local situations. These papers are revisions of oral presen­
tations from a panel at the 11th annual meeting of the Middle East Studies
Association, November 10 1977, in New York City. We appreciate the in­
vitation by the Afghanistan Council to publish them together as an Occasional
Paper of the Afghanistan Council of The Asia Society. This format preserves
the unity in which they were conceived and reaches those most immediately
concerned with Afghan studies.

A loose comparative frame for examining the diversity of ethnic processes

and intergroup relations in contemporary Afghanistan is provided by the
historic Pashtun expansion into the territories and activities of other
groups in Afghanistan during the present century. The idea for papers on
this topic, orioginally suggested by Richard Strand, grew out of discussions
among the authors and others beginning with a conference on development in
Afghanistan at the University of Nebraska-Omaha Center for Afghanistan
Studies in September 1976. The particular researches upon which the present
:papers draw were not originally designed with this topic in mind. Barfield's
research in Qataghan focused on the organization of contemporary pastoralism.
Shahrani's work in Hakhan and Pamir dealt with cultural ecology. Strand's
interests in Nuristan were primarily linguistic and ethnohistorical. Can­
field has published substantial and original contributions on the organization
of sectarianism among the Hazara. Anderson's Pashtun research focused on
tribal law' and interpersonal relations. But each of us worked across
ethnically plural settings and, in the course of separate researches between
1966 and 1974, had reason to take notice of the coincidence of Pashtun ex­
pansion and national consolidation in those settings.

Given this overlap we chose a broad common focus on the impact on local

minorities of this coincidence. Additional contributions from researchers

in Europe engaged in work more directly related than our own to "ethnic"

problems were precluded by distance and the time constraints of the MESA

program format; so rather than attempt comprehensive coverage we have

opted for working papers with limited aims in the conviction that the primary

need in Afghan studies is for a richer descriptive base. Anderson's intro­

ductory remarks to the session are expanded here to indicate some of the

context that these papers address.

That introduction as well as the present papers benefit from discussion at

the meeting by Warren Swidler, Associate Professor and Chairman of Anthropology
at Brooklyn College-CUNY, and Karen Blu, Assistant Professor of Anthropology
at New York University. Swidler has conducted extensive research on ecology
and social organization in Baluchistan (Swidler 1973), and Blu's research

i
among North American Indians has centered on questions of ethnicity in
complex societies. For economy and because our intentions here are more In
descriptive than theoretical we have chosen to take note of their com­
ments and of those from the audience in the present drafts rather than
to record them separately. Thus it is all the more fitting that the
stimulation of their remarks be especially acknowledged.

Questions raised by and in response to these papers make the necessity In I

for more intensive descriptions of actual local settings in Afghanistan Car:
all the more compelling. These settings are the contexts in which the shi]
lives of all the people there take shape and toward whose comprehension spel
les ~
these working papers are offered. Opinions expressed in these papers
are those of their authors and do not reflect those of t .heir institutions, pha.
funding agencies, the Hiddle East Studies Association or The Asia Society. of 1
rel Ci
chr
.TI-lA slo
RFS Sou
socl
June 1978 of t!

In t
nine
them
The
tend
as s
(195
Groe
seco
its
for
aski
expl
or,

Ques
have
cone
Greg
some
Farh
pers,
part
on t
find
abou
noti
whic
orga
ethn
some

ii
 in
more Introduction and Overview
:::om­
han Jon W. Anderson
e University of North Carolina

'ty
In characterizing the Near East as a "mosaic" of cultures and peoples
stan Carleton Coon (1958) sought to capture a particular part-whole relation­

the ship betvleen small, diversely originated, often endogamous and economically
ion specialized local groups and the embracive polities that seem something
s less than the sum of their parts. His descriptive met. aphor broadly em­

tutions, phasizes the perdurability of local di versi ties in t hf ';omposi te societies

ociety. of this region and draws attention to an organizatioL ,.: haracter in the
relations among its various peoples more complex th o: t suggested by
chronicles of conquests and successions. The variet> peoples along the
slopes of the Hindu Kush where the great culture are,; :. )~: Central, South and
Southwest Asia meet in Afghanistan exemplifies this e no '.l.:cance of local
social and cultural forms within groupings done and undo ne over the course
of time that never quite transcended the diversity of their parts.

In the literature on Afghanistan attention to this diversity keynotes

nineteenth century accounts of the variety of peoples, comparisons between
them and often ingenious speculation about their relations and origins.
The explanatory attention embedded in these descriptions and reconstructions
tended to focus on ethnogenesis and a kind of cultural inventory analysis
as sufficiently accounting for the diversity itself. Bellew (1891), Bacon
(1951, 1958), Schurman (1962) f Janata (1962/63, 1971), Franz (1972),
Groetzbach (1972) and especially the many composite handbooks and other
secondary treatments offer data and interpretation in such a frame. Given
its classificatory emphasis this approach tends to take diversity itself
for granted, explaining it by reference to something else, rather than
asking what sort of social "fact" it is. It tends to accept rather than to
explore, except in historical terms, the self-identification of local groups
or, as is frequently the case, how their neighbors identify them.

Questions about ethnic diversity and the relations between different groups
have also been put in terms of modern, but still essentially exteriorist,
concerns with the integration of multi-ethnic states. Adamec (1967),
Gregorian (1969), Ne,.,rell (1972), Poullada (1973), Dupree (1964, 1973, 1975),
some Russians (Slobin & Slobin 1969) and several Afghans themselves (~,~.,
Farhadi 1970, Kakar 1971, Miran 1977) essay on ethnic diversity from the
perspective of relating modern, national government to "traditional" society,
particularly in respect to political and economic development. Depending
on the authors' point of view and professional interests they variously
find problems and prospects in the ways they perceive diversity. Questions
about conflict often frame such discussions, for embedded in them is a
notion that "traditions" such as ethnicity obstruct "modernization," by
which seems to be meant the sort of politically unified and economically
organized society such as Western nations are imagined to be. In this frame
ethnic relations are seen through policy problems that focus on explaining
something else.

1
Between reduction to historical residues and over-construction as a policy
problem, the dynamics of ethnic and intergroup relations in Afghanistan
have received on the whole little first-order description. With a few
notable exceptions (Canfield 1973a & b, Ferdinand 1962 & 1969, Barth 1959 &
1969, Tapper 1973, Glatzer 1977), there is in print very little case
material of much depth upon which to g.l:'ound analysis. Instead, information
apart from that based on archival research (~.~., Gregorian 1969, Kakar 1971,
Groetzbach 1972, Tapper 1973) tends to be folklcre-like, an oral body of
knowledge in circulation among Afghanists that derives from their personal
experiences, often as not largely in Kabul, and reflects concerns of Afghans
at the center and top who aretbe natural acquaintances of historians and
political scientists. Anecdotal information is occasionally put imaginatively
to use (~'5L" Poullada 1973); but in a perspective at once ethnographically
wider and more sociological, it largely formalizes situations and concerns
peculiar to the more national-minded and foreign-oriented sectors of the
society. The reference of such views, no less than their context, is
necessarily limited by being drawn from a narrow base.

Both to expand the empirical base and to relate such discussions to the local
l .evels where people from different groups meet and interact in everyday
affairs, the present papers ex a mine specific case materials drawn from recent
ethnographic research in rural Afghanistan. lIS their primary intent is
descriptive the frame of these papers is open and exploratory. They do have
a common substantive focus, hmvever, on contexts set by the historic Pashtun
expansion throughout Afghanistan during this century.

* * *
The character of contemporary ethnic diversity in Afghanistan and the related
problems of national integration are set locally in large measure by the
expansion of Pashtun settlement and influence beyond their historic center in
southern and eastern Afghanistan (and the North-West Frontier province of
western Pakistan) " The reasons for this expansion are a complex mix of mostly
external factors that both fixed current international boundaries and re­
directed Pashtun attention from its historical focus on the Indus valley
toward what is now thB interior of modern Afghanistan. The history of the ·
past two hundred years is not only one of fitful ascendence of Pashtun on a
stage not of their own making. Since the nation-building reign of Amir Abdur
Rahman (1880-1901), the on-the-ground reality in many parts of Afghanistan
has included Pashtun immigration (see esp. Ferdinand 1962, Kakar 1971, Tapper
1973). Many aspects of ethnic and intergroup relations currently are con­
sequences of these processes and of local awareness of them on the part of
non-Pashtun.

Kakar's and Tapper's historical inve stigations and our own field researches
suggest that this expansion is not simply a matter of conquest, certainly not
since Amir Abdur Rahman's decisive incorporation of the Hazarajat, Turkistan
and Nuristan before 1900. These papers point to a more multifaceted and
evolving variety of interfaces, including those involving Pashtun only
partially (~_~., Strand's and Shahrani's papers in this collection). In out­
lining features of specific local situations these papers make no attempt
to be comprehensive or conclusory about Afghanistan in its entirety. They

2
 are instead working papers that aim at once more narrowly and more
wholistically to elucidate organizational dimensions of what, following
the initial suggestion of Richard Strand, was characterized for convenience
as "ethnic processes and intergroup relations."
&
The term "ethnic" covers a variety of reference, both substantive and
)n methodological. Karen Blu noted that sociological concepts of "ethnicity"
371, contain embedded in them our particular concerns about nationality and may
be inappropriate in other cultures Ca point also raised by Canfield, below)
1 and ~"'arren Swidler discussed how satisfactorily to identify communi ties for
oms analysis. Problems of predefinition and reference are encow1tered, for ex­
ample, in trying to relate realities on the ground to "ethnic group" maps
ively of Afghanistan, of which that published in Sovetskaya Etnografiya in 1955
ly (reprinted in' the Central Asian Review in 1956) is the base model for all
s subsequent maps. Concerns of this sort were one of our points of departure,
for the true complexities of population distribution are obscured rather than
revealed by such representations. l The simple fact that bOlli1daries are not
all of a piece, tha:t they vary according to the situation, is all the more
glossed over the more detailed such maps become. It is to retrieve a fuller
ocal and more organizational grasp of the dimensions of "ethnic" identities and
their role in intergroup relations that we emphasize here process or the
cent action-oriented properties and impacts of intergroup relations in specific
cases. Those processes framing intergroup relations, rather than "ethnicity"
have generally in Afghanistan, are our immediate topic .
.tun
By the tenn "ethnic process" I unders·t.Ood Strand to mean - and others seem to
to as well - that "ethnic" identities sociologically emerge in situations where
people of different traditions and organizations come together or are brought
together in contexts set by terms external to themselves. Indeed, it is the
.ated emergence of wider community, such as in a territorial state, that Cl'eates
\'ethnicity" among heretofore disjoined groups which have become components in
:r in some encompassing structure.

lostly The encompassing structure in this case is constructed of national suzerainty

and Pashtun population expansion in concert with the rise of that suzerainty.
The key figure connecting these is Amir Abdur Rahman who, to consolidate his
.e realm, encouraged Pashtun emigration to other parts of Afghanistan as no ruler
. a had before (see Kakar 1971, and for case studies Ferdinand 1962, Tapper 1973).

ibdur
To be sure, the "Iron Amir" 'vas not slow to treat homeland Pashtun as just
.n another component of the nation; much to their annoyance; but essential to the
.pper policy established by him and followed by most subsequent rulers was the
joining of nation-building to Pashtml resettlement outside the hills and valleys
,f from Nangarhar through Kandahar and nearly to Herat that is their t~aditional
homeland (watan). National integration was not the sole purpose of encouraging
emigration, and colonization is far from the whole story, but the connection
~ es between national consolidation and "Pashtunization" endured, especially in

, not
the consciousnesses of non-Pashtun.

,tan

out­ IFor an examination of the orsanizational terms of more specific population dis­
lt tributions, see Anderson (1975) for Ghilzai; comparable material is provided
y by Barth (1959) for Swat, Steul (1973) for Khost and Glatzer (1977) for Pashtun
in the Firuzkoh.

3
This is not to interpret the Kabul government in the twentieth century as a
creature of Paahtun tribalism. The government necessarily and naturally has
interests of its own and a nationalist constituency. And the feelings of
many Pashtun tribesmen toward the government differ little from those of non­
Pashtun. All alike resent interference and seek advantage from Kabul. But
Pashtun are often perceived as enjoying favor. Many provincial government
posts and until recently nearly all army commissions were filled by Pashtun,
and Pashtun figure prominently among the beneficiaries of such population re­
distribution schemes as nomad settlement and agricultural development, all of w~
conduce to impressions on the part of many that Pashtun and government in­
terests coincide. The "Iron Amir" was clear enough about his own motives in
his memoirs (1900). Removing Pashtun, often in whole groups, to other areas
by force or persuasion had the dual effect of converting potential dissidents
into national partisans and thereby fostering the new community of a national,
if still plural, state. Results, of course, are matters of degree that vary
from place to place; but broadly that effect was, in part, as expected.

As immigrant Pashtun develop relations with other local people, much of the
feeling on both sides depends on the nature of those relations and the auspices
under which they develop. The contrasting situations described by Barfield
and Shahrani (and to a lesser extent by Strand) especially make this clear.
Echoing Canfield's earlier study (1973a), Shahrani and Strand describe situations
of partial displacement in which local groups are penetrated and made into
specialized components of a larger system at least indirectly fostered by the
government and over which the local people have little or no control. Contraril;
where a more or less open country was "developed" jointly by members of many
groups responding individually in the context set by Spinzar Company activities
Qataghan, Barfield finds the emerging status system based less on origins or
group modalities than on socio-economic relations organized along more class­
like lines with rich/poor and employer/employee being the salient distinctions.
Moreover, the generation of ethnicity from kinship is patent: where kinship
is not organizationally relevant as the basis of grouping (on at least one
sidel, neither is "ethnicity."

It will be interesting to see if the researches of Asen Balikci in Narin, of

Pierre & Michelene Centlivres in Baghlan and Takhar, of Richard & Nancy Tapper
in Pariab and of others more directly concerned with "ethnic" problems will
confirm the duality of these patterns. If they do, it should provide one
possible key to the complexity of intergroup relations in Afghanistan and to
the contradictions of fragmentary reports. A rough rule-of-thumb would relate
penetration or encapsulation of local groups, such as Shahrani describes in
Wakhan and Strand in Nuristan, to heightened (or engendered) self- and other­
consciousness in terms of making boundaries the basis for organizing activities;
while frontier (unbounded) situations where all are immigrants in a cowmercial
context that fosters typing in functional terms, such as Barfield describes,
tend to be non-ethnically (or less ethnically) conceived class-like phenomena.
Such a rule does not predict the relations of structural to functional
variables but identifies what about those relations is problematlc. Certainly,
these papers do not support viewing such processes as mutually exclusive
or as exhausting all possibilities, but rather that these processes are com­
plexly interwoven according to how particular situations get defined. Also
certainly, the situation in Kunduz, where the Spinzar enterprises attracted

4
s a
has
population into a previously empty but developable area, is unique in rural

f
Afghanistan. vTnat the papers do suggest is that no single variable or

non­
simple formula will describe all situations or even any particular one.

But
nt
These four papers provide insufficient basis for any conclusions except to
tun, urge that intergroup relations are complex and emergent, that they are tied
n re­
to specific local conditions and that they change even as they are encountered.
11 of
Sociological classification alone will go only so far in grasping this com­
n­
plexity before merely reproducing the variety it aims to encompass. It is
5 in
the particular lesson of Canfield's present paper that we really know little
reas
of how (any) Afghans think of ethnicity and intergroup relations and that
:3.ents
this is crucial for grasping what is "actually" going on. In a sense, the
ional,
other papers and the panel as a whole have taken shape in the context of
'Vary
Canfield's O\vn prior pioneering research (1973a, 1973b) on the cultural­
ecological aspects of marginal sectariani s m. His present contribution goes
beyond that frame to truly new questions in Afghan studies. Whether his
::he
specific argument that secta rian difference s loom largest in Afghanistan is
lspices
born out - and the present precis conta ins only a portion of his discussion ­
~ld
it is indubitable that the more profound grasp of intergroup relations will
'ir.
corne of apprehending how Afghans the mselves conceive of and represent them.
Ltuations
That is largely unknown at the present.
:0
, the
mtrarily, * * *
[any
In this connection, information from the Pashtun side of the encounters that
:ivities in
are our subject is a neccessary complement to the present papers. I have
) or
suggested elsewhere (Anderson 1975, -1976) that Pas htun tend to think of kin­
_ass­
ship (intluding affinal as well as descent relations) as the mediator of
:tions.
individual and collectivity, or that ,,'hat we call kinship and what we call
lip
ethnicity are for them joined in a single framework. Pashtun conceive of
le
thems e lves as united and bounded by common patrilineal descent; so "ethnicity"
presents itself in some of the same forms as kinship. "Ethnicity" and "kin­
ship" are covered by the same term, gawm, implying co-descent (which seems to
of be the case also for Nuristani and Turkic peoples, cf' J Shahrani B.nd Strand in
rapper
this volume). The crucial relation in that frame is between the culturally
[ill defined sameness or difference of social actors. On the basis of sameness of
identity, cooperation and joint actions can proceed in a frame of mutuality
l to
or equality, a collapsing of distinctions b e tween persons. On the basis of
'elate
different identities, only complementarity can organize a relationship, and
in complementarity for Pashtun fosters not mutuality but opposition resolved as
.her­
hierarchy. Put abstractly, this comes near to being true of social inter­
vities;
action generally; but it is culturally true for Pashtun. The terms of the
Ircial
symbolic frame in which they conceptualize social relations can be seen con­
es, cretely in the variation of sharecropping arrangements from those interpreted
ena.
and enacted as partnerships to those organized as between employer and em­
[ ployee. This ontology holds the implication for Pashtun, quite (perhaps too)
a.inly, bluntly, of relationships marked by superiority/inferiority entailing the
superior encompassing the inferior in his own purposes and ide ntity. Thus
om­
"ethnicity" as a mode for relating to non-Pashtun takes shape in a hierarchical
so . frame. 2
ed
2In this connection, the research of Frederick Barth in Swat and Baluchistan
in West Pakistan (1959, 1960, 1969) has a relevance so far little exploited

 Difference (hierarchy) can imply enmity, though not neces3arily. What it

does for Pashtun is to pose dilemmas of dominating or being dominated, the
only alternative being to break off relations entirely, which is frequently
the case. From this way of thinking derives much of the storied aggressiveness
and self-assertiveness of Pashtun demeanor, manners with which others are
sufficiently familiar to resent on occasion and often (in private) to satirize.
It is as well reflected in the British literature in the ubiquitous descriptions
of Pashtun "character" as treacherous, changeable, aggressive, proud, indepen­
dent, unscrupulous and the like. Non-Pashtun are acutely aware of this inter­
actional pose, encountering it most often in those Pashtun whose relations
w:it:.h everyone are complementary, the svTaggering and often combative - but
equally hospital and flexible - nomads.

Such behavior and others' reactions to it should not be taken prima facie as
evidence of hostility. Hostility is understood by Pashtun and non-Pashtun
in their own terms and has many roots and expressions, of which hierarchy is
not the most important or essential, as anyone can attest who has experienced
the very hierarchical and arms-length relationship of Pashtun hospitality.
But for intergroup relations in which Pashtun figure qu~ Pashtun this feature
of their interaction styles relates to phenomena that many of us have ob­
served privately and to \'lhich many Afghans draw our attention - namely, that
with the Pashtun expansion has gone a spreading "Pashtunization" of public
manners. As one rl'aj ik near Kabul put it to me: "How else is one to d8al with
them [except in their own terms]?"

If it be the case that Pashtun put the stamp of their own styles on their
relations with others - and historical evidence is ambiguous in this regard ­
it would be fruitless to ask to what extent. The important fact is that the
impression obtains among at least some Afghans and, entering into their
definitions of situation, is thereby part of those situations. Of COi.lrse,
such models of demeanor are stereotypes; but stereotypes are "real" enough
representations to the people who use them to comprehend their social worlds
and who regard them as of those worlds. Analytically, this impression of
"Pashtunization" points to the emerging culture or, more likely, cultures of
intergroup relations that accorc1pany articulations of heretofore more separate
local communities into a wider community as components of that larger whole.
Consisting of conceptions of self and others by which such relations are
locally understood as coherent activities, cultures of intergroup relations
combine pre-existing conceptions with newly generated ones in complex ,'lays and .
various settings that are little understood but highly suggestive.

Whether it mayor may not be historically the case nhat the behavior of others

by non-anthropological area specialists. In particular, aspects of the impact

of ethnic boundaries on local organization that Barth described for Swat seem
to be r.eplicated in parts of Nuristan and the Hazarajat where PashtW1 absorb
nor.-Pashtun to regional (but not always Pashtun) systems as a political
reflex of acquiring the basic asset of agricultural society, control of land
(cf., Ferdinand 1962). Pashtun success qua Pashtun depends on this (ct.,
Barfield, below).

6
 -'--- - ­

't is being "Pashtunized," and irrespective of how much others contribute their
he own styles to the organization of intergroup relations, many Afghans commonly
tly attribute this to Pashtun and its diffusion to what has been characterized in
iveness other connections as "internal imperialism." Infonnants indicate that one
'e aspect of this diffusion is that in addition to marking relations of Pashtun
tirize. nomads with others - pastoralist or otherwise, Pashtun outside their home­
rip'tions land have many characteristics in common (cf., Anderson 1975: 582, 585) - it
depen­ also characterizes the pose adopted by many government officials in provincial
'n'ter­ posts. There, as with nomads and other "strangers" not on their O\.;n home
ns ground, it may be a reflex of situated hierarchy and the comple.mentarity be­
tween autochthone and outsider, especially in the petitioner-disposer relation­
ship that characterizes most persons' dealings with government in the country­
side. Certainly, not all officials seek consistently so to present themselves,
~ as
but the powerful stranger is still identified for many with Pashtun and with
.un impress.ions of their expansion and influence paralleling that of the government .
y is
.enced The specific "Pashtunization" to which infonnants refer is probably less a
:Yo matter of copying Pashtun styles than of the fact that a ubiquitous feature
~ature of intergroup relations is that they involve dealing with Pashtun and that
;­ Pash.tun approach such relations in a Hobbesian manner. Pashtun are in the
'tha't first instance the most (though not the only) mobile element in Afghanistan
.ic and, to many, the "stranger" who is par excellence both insider and outsider
. wi'th (cf., Simmel 1950: 402-408). That mobility has been exploited by the "ruling
institution" since the Iron Amir made it his instrument of national consolidation,
so a culture of intergroup relations importantly takes shape around Pashtun
.r (immigrant) activities and in response to those activities. Moreover, the
rard ­ per definition complimentarity involved in such relations and the ways that
: 'the Pashtun gloss them tend to encode a hierarchical construction on intergroup
relations. To have any other requires a gloss of mutuality which mayor may not
;e, emerge according to the specific activities in which intergroup relations are
19h organized. If Barfield's analysis is any guide, then transcendence of pre­
)rlds given modalities for organization (such as qawm) could be said to involve the
)f entrance of all into the social field as "strangers," as equally from outside.
While situations do evolve, as Shahrani and Strand make plain, the auspices
-~s of under which they are inaugurated tend to survive reflexively in adaptations
)ara'te
lOle. to them. The respective comrnerces which organize Xakhar and ~vakhan-Pamir thus
set the situations to which people there respond .
.ons
lyS and i n Two very general levels of intergroup relations can be seen as dialectically
related in activities taking place in particular situations that emerge from
the participants' attempts to comprehend them and their roles in them. If
others indeed as the papers suggest, such relations turn out to be organized by
activity-focused cultures, then that may be one of the two most important things
to grasp about them. The other is that these "cultures," if they may be properly
impact so-called, center on the role of the "stranger," whose position both inside and
seem outside, implying deep if narrow familiarity, was located by Simmel (1950)
tbsorb quintessentially in the trader and the financier, participants radically un­
committed to and hence "objective" about the values of the other in the sense of
land regarding those as something to know about rather than to believe in.

The ambiguities and seeming contradictions of intergroup relations inhere in

this dual character of the Stranger:

. 7
 ••. strangeness is not due to different and ununderstandable
matters. It is rather caused by the fact that similarity,
harmony, and nearness are accompanied by the feeling that
they are not really the unique property of this particular
relationship: they are something more general, something
which potentially prevails between the partners and an in­
determinate number of others, and therefore gives the re­
lation, which alone [of all possible relations] was realized,
no inner and exclusive necessity.

On the other hand, there is a kind of 'strangeness' that

rejects the very commonness based on something more general
which embraces the parties. The relation of the Greeks to
the Barbarians is perhaps typical here, as are all cases in
which it is precisely general attributes, felt to be specif­
ically and purely human, that _are disallowed to the other.
But 'stranger,' here, has no positive meaning; the relation
to him is a non-relation; he is not what is relevant here, a
member of the group.

As a group member, rather, he is near and far at the same time,

as is characteristic of relations founded onlyon generall-y-­
human commonness. But bet\veen nearness and distance there
arises a specific tension when the consciousness that only the
quite. general is common, stresses that which is not common .
... For this reason, strangers are not really conceived as in­
dividuals, but as strangers of a particular type: the element
of distance is no less general in regard to them than the
element of nearness (Simmel 1950: 407).

Simmel thus urges us to grasp the transcendence of particulari ty (~.5I"

as described by Barfield) as well as its reification (~'5I" described
by Canfield) and operationalization(~.~., described by Shahrani and
Strand) within a single, unified frame whose dimensions are fundamental
social processes.
* * *
These are only some of the interesting avenues suggested by more empirical
constructions on large abstractions that have quite human referents. A
fuller grasp of ethnic process and intergroup relations in Afghanistan will
require not only more ethnography but historical study, incorporation of the
growing body of population and other "development" research and contribution~
particularly from Afghans themselves. If some formulations in the past
have been anthropologically naive, much of the reason must lie in the want
of systematic ethnography. That concern, remarked on by the discussants,
is apposite in these papers: a general need for a more thorough grasp of
the communities involved or the stages on which actions take place and a
systematic grasp of the actors' conceptions of these matters on a much more
fundamental (ontological) level. The authors were and are aware of such
problems. As anthropologists, we are all co~~itted to elucidating Afghan
realities rather than simply examining those from the perspective of our oym;
and the very nature of anthropological study is intensive inquiry. Our
aims in these papers were not to resolve questions but to identify them for,
in th.e state of this issue, we are still discovering the right questions.

8
 Ethnie Competition and Tribal Schism in Eastern Nuristan

Richard F. Strand
Research Technology, Inc.

The protagonists in this study are the Kom Nuristani people who occupy
the lower LaQ~ay Sin basin and areas across the Kunar River in the
easterDl110st Hindu-Kush region of Afghanistan, and a loose confederation
of Gujar and Meswani patrilineages whose members have established
residence in Kom territory in various side valleys of the Kunar, The
Gujars are an Indo-Aryan-speaking people who have spread west';.'ard from
the Gujrat area of the Panj<ili. The Heswar;i are Pashto speakers but
they deny that they are "Afghans" in the sense that they do not fit
into the Afghan [Pashtun - ed.] tribal genealogical charter. They are
also called in Pashto ban9awal, 'alpine-stable owners,.l Although the
Gujars and Heswar.1i are ethnically and linguistically distinct from each
other, the Kom often deal with them collectively, and I shall refer to
them here by the term Gujirbandevol, which means in Kamv{ri (the Kom
language) literally 'Gujars and banSlawals'.

According to Nuristani traditions, Afghans long ago gained most of the

best bottom land in the Kunar-Kabul basin for cultivation forcing the
Nurist:anis into the back valleys and mountains where cuI tivahle land is
scarce. The Nuristanis have managed to subsist in such areas because
they have a truly mixed economy of agriculture and animal husbandry and
the scarcity of productive agricultural land is offset by the abundant
dairy production afforded by good alpine pasturage (Strand 1975). Since
the Nuristani occupation of the Hindu Kush recesses, interethnic territorial
conflicts have centered over control and use of such pasturage. 2
1 Before 1896 such conflict was usually confined among the various Nuristani
tribes; but since that time, when the Nuristanis were conquered, converted
to Islam and incorporated into the Afghan state, they have more or less
desisted from intertribal warfare and tribal boundaries have become
ieal fixed.
A
will
of the
butions This paper is based on field research in Afghanistan during 196:-1969 and
t 1973-1974. Funding came partially from the Henner-Gren Foundatlon for. .
,.rant Anthropological Research) Inc. and the South Asia Program of Cornell Unlverslty.
ts,
of IAn account of the ethnolinguistic position of the Nuristani peoples
a appears in Strand 1973.
more
:h 21 do not consider the slaughter that the Nuristanis wreaked on the
lan neighboring Afghans before 1896 as territorial conflict. Rather, it 'das
Ir own; institutionalized vengeance for the Afghans' murder of the Korn culture
hero Gis, an act that symbolized the expUlsion of the Kom from their
1 for, originai homeland in the Kunar basin.
IS.
However, in the past 70 years the Gujars and the Me~wapi, both of whom
subsist almost exclusively on alpine goatherding, have responded to
demographic pressure from the east by infiltrating the southern pasture­
lands of the Kom tribe. Because the Kom herds had not recovered from
the decimation suffered during the conquest of 1896, those pas~~relan?s
had fallen into disuse, and the Kom had agreed to allow the Gujlrb~?evols
to use the area seasonally in return for a grazing fee of one goat per
twenty grazed. The agreement worked fairly well until about thirty years
ago, when the Gu:i'{~bar:?evols reneged on the grazing fee and began estab­
lishing permanent - ~ettlements in Kom territory. Si!lce then hostilities
have become perennial with occasional shootings and rustlings on both
sides.

The Kom have always had a reputation as the "wildest" of the Nuristani
tribes and in preconquest times they could have easily expelled the
GUJ{rbaQ~evols from their territory. Yet the Guj'irbar:Slcvols remain suc­
cessful in achieving their designs on Kom territory. How have they
succe8ded?

First, they have secured the persistent backing of key provincial officials.
Having enjoined the Kom from waging open ,varfare on the intruders, these
officials have pursued a negotiated settl e ment to the issue, with them­
selves as mediators i but all compromi ses have so far favored the GUJ'i'r­
bar:?evols' position. The Kom allege that the GU] irbal!?evols have sometimes
bought th~ backing of certain officials and that feelings of e1::hnic
solidarity between Afghan provincial officials and the Hes\jiqli foster an
anti-Nuristani bias. However, the reason for the government's failure
to produce a solution that is accept a ble to the Kom apparently goes beyond
considerations of prejudice or intere~hnic intrigue. An account of events
that happened in the early 1970s illustrates rather typically how the re­
peated attempts of the Kom to settle the issue through governmental
mediation are continually thwarted. 3

After some skirmishes between KOHl and Gujars, the wali (governor) of
Kunarha: Province requested tribal leaders from both sides to gather in
the provincial capital, Chaghan Saray (Asadabad). There he conducted an
inquiry at which the Kom explained that the Gujars had continued to renege
on their agreement to pay a grazing fee for use of Kom pastureland. To
substantiate their claims the Kom produced \'lritten contracts and official
documents that defined both the agreement between Kom and Gujars and the
territorial rights of the Kom tribe. The Gujars countered that the Kom
had denied them access to the pastureland but they hedged when asked
whether they had reneged on the agreement first. The wali decided to
draw up a new contract reiterating the terms of the original agreeme nt;
but before he could do so the Gujar representatives fled Chaghan Saray
in the middle of the night without lending themselves to a new agree­
ment.

3 This account comes primarily from one Kom leader \-'ho was intimately in­
volved in the affair. As such, it may contain some distortion of the
actual events, but the form in which it appears is ethnographically valid
in that it expresses the belief of a majority of the Kom.

10
 The Kom leaders then petitioned King Zaher Shah directly in Kabul. The
king ordered th€ wali to round up the recalcitrant Gujar leaders and ac­
company them to Kabul where negotiations were to continue. However, be­
.re­ fore the Kom leaders were called in, the king (they allege) met secretly 4
with the wali and worked out a solution. Being partial to the Nuristanis
ds
the king supposedly ordered the wali to offer the Kom some land in the
levols
canal project near Jalalabad as compensation for their troubles and to
deport the entire Gujar population to Turkestan where he would give them
'ears land for resettlement. 5
;tab­
The willi then requested the Kom to give him time to consider the matter
.es whereupon he Vlould inform them of his decision. After a few days he
announced that the king would give the Kom 600 jeribs of land at the
canal project and that all Gujars living on Kom la.nds in the Kunar valley
would have too leave, but that Guj ars living on Kom land near the village
of Dungal could remain there indefinitely without paying a grazing fee.
Interestingly, the latter group of Gujars live in contact with a group of
;uc­ l,:1eswal}.i in an are a that stands on the route of -transhumance between the
Kunar and the high pastures that are the main area of controversy. It is pre­
cisely this group of GJ-'j/irbar:l¢!e'vols that pose the most acute threat to Kom
territorial integrity. The Kom strongly rejected the \'l~ni's decision,
:icials. saying, in th e \'lOrds of one Kom leader, that "un-i :.il all the Gujars are sur­
~ese gically excised to the core, we Kom \"i11 not see a cure."
,m­
,r­ There is a strong feeling among many Kom leaders Ulat the ruling estab­
,times lishment of Afghanis tan pursues G_ policy of di vide-and-'conquer toward the
Nuristani tribes so that they may keep tribal unity weak and thereby
: an govern with ease. The Kom see proposed sett1emec'1ts of the kind just
:e illustrated as thinly disguised attempts to erode their terri'corial base
leyond by sanctioning an enemy in their midst.
,vents
, re­ A second reason for the Gu::ri-rban~evols' success is that -they have effec­
tively bough'c off an important minority of Kom l-2aders, perpetrating a
schism betvleen the latter and the ma jority of KO;11 -tribesmen. The sellout
\Vas effected in various ways. Some leaders, it is alleged, were bribed
outright. Such allegations are almost impossible to prove but the apparently
_n unmoti va ted magnanimity of E>uch leaders toward tlwir supposed enemies
l an tends to reinforce suspicions. Other leaders o",n land close to Guj'{rbandevol­
~enege infil trated areas and have become economically d,":pendent of them for their
To herding activities (Strand 1975: 131-132). ThoE8 Kom also fear reprioals
_cial on their poorly protected holdings if hos -tili ty beble e n Kom and GUJ{rbar:s3-evols
the were to escalate. In some remote Kom settlements east of the Kunar River
:om

4Regardless of any ill feelings that they may: have toward the government in
It; . general, most Kom held the king in high regarCi. 'I'here is even a myth
ly current in Nuristan that ~aher Shah was actually illegitimately fathered
by a prominent Nuristani general, making him a kind of clandestine kinsman
of all Nuristanis.

5The canal project, built under Russian technicaIL assistance to irriga'te

in­ parts of Nangrahar Province, has provided Nurista;:ris with a major source

of emplo~nent outside tribal territory. The propmsed aLea of resettle­

'alid ment in Turkestan is in Qataghan Province wher e nany }\fghans from eastern

Afghanistan have recently been given lemd. [~ ..' ).lfarfield I s paper in this

volume] .

11
many of th e inhabitants have intermarri e d with Gujars creating affinal
and matrilateral kinship obligations toward the Gujars. Finally, a few
powerful Kom have made clandestine -de als to collect privately a r e duced
grazing fee from certain Gujlrba0~evols allowing the latte r to remain
unmolested within Kom territory.

For differing reasons these tribesmen stand to lose economically j,f the
Gujars are expelled from Kom territory. Th e y have been collectively labelec
the guJ'ir c;1ala (' Su-j a r faction') by their more numerous opponen t. s (wh o m
I shall call the traditionalist faction, led primarily by the represen­
tatives mentioned in the narrative a bove), and they have succeede d in
subverting most conc e rted a ction against the Guj'.l.rbanc;18vols. They often
do so by c..isrupting the community confe rence s at whi c h collective tribal
decisions are made so t ha t th e confere nc e s bre a k up be f ore d e cisions can
be r e ached. As a result, no unified s t rateg y a_go.inst the intrude rs can b e
maintaine d and any r e pris a ls ag a inst the Gujlrbal}¢\evol s are usually the
work of indi v idual s rath e r than of the cornmunity as a whol e .

How has the Gujar faction been able to ma i ntain a pos i-tion that pote ntial­
ly threatens the terri torial inVc~gri ty of the whole t.ribe? vlhy have th e y
not be e n r e pudi a te d and ostra c ized by the majority o f tribesme n? Answ e rs
to thes e que stions require a bri e f look a t some fund ame nt a l principl e s
und e rlying Kom s ociety.

To preve nt tribal disunity in a pre c10rnin a ntly hostil e intere thnic environ­
ment, the Kom evolve d a n e ffective sys -t em o f s ocial cohesion emb odied in
the principle s of citizens h ip a n d kin s hip. As t he Kom def i ne th e se prin­
ciples, 'citize nship ' (ImgI'tfmvor) is t he s e t of righ -ts and obligations
incumb e nt on -th e r e side~~s of a:--conmmn i ty (gI'om), a nd I kins hip ' (jatrevor)
is the s e t of rights a n d obliga-t ions incumb en t on p e rsons who trace a
rel a tionship to a commo n a n ce stor. I have discus se d t_hc::s e principl e s
el s ewhere (S t:ca Lid 1 9 74 a. and b) an d will no t elaborate on them h e re. Suf­
;!;ice i t to s a y th a t. app e21 s to t hese princ i p les, backed up by sanctions
of os -tr a cism from cormnunity cmd kin, usually have b e en suf f icient to
prevent individual s from gros s ly contrave ning tribal intere sts.

Howe ver, the a c cel e r a ting e x p osnre of the Ko m to the rules and va.lues of
the Afgh a n st a t e incr ease s the potenti a l for indivi dual s to diverge suc­
cessfully from traditional conce pt s o f citizenship and kinship. Tribesmen
have the option o f invoking either t r aditional or national values in
purs uing their economic and political go a ls and they can find s an ction for
their actions through either t raditional tribal means or through the Afghan­
Islamic legal system. ll. stri k ing e xampl e of the effect of such options on
Kom social structure _ts the compl e te br e akdmvn of lin e age e xogamy.
Traditiona lly tribe smen were prohibite d f rom marrying wome n of the ir own
patrilin e ages under p e nCllty of oSL~acism from the lineage. Howeve r, the
potential for extracting a lower bride price from an agnate 'das exploite d
as soon as some tribesme n felt confident that such action could b e sup­
ported by I s lamic law which lacks a ny prohibi t.ion of agnat ic marriag e . Toc:
after three g e nerations of Islamicization, agnatic relationship is hardly
a consideration in the choice ofa spouse.

,'">
.1..":'
 .~-- _.-­

,al

The leaders of the Gujar faction use such alternative Islamic and
fe\¥

nationalistic values to legitimize their position. They argue that the

,ced

Gujars are essentially poor men and that the Kom should show benevolence
,n
towards their less fortunate .l'luslim broth e rs. They say that nowadays
the whol e country is unifonnly Afghanistan and no longer divided into
a number of indep e nd ent trib a l territories; th e refore, because the Kom
the
and the Gu1{rba~~~vols are all citizens of Afghanistan, and becaus e there
, labeled
is an excess of pastureland in the disputed a rea, the Kom have a pat­
:whom
riotic duty to share their traditional land with th e ir compatriots (~~tal)­
;en­
~~rs). These leaders h ave es se ntially enlarged the tra ditional sense of
.n
' community' to include the community of Islam and the state of Afghanistan
)ften
and they strongly imply that a stance against their position is e ither
~ibal
heretical or trea sonous.
, can
can be
These a rguments have pe r s uade d a significant numb e r of non-politicize d
the
trib esm8 n to lend a t l east passive suppo:ct to the Gujar faction so that
th e p ol i·tical leaders in the Gujar faction command a disproportion ate ly
large ba.cking o f trib e sm e n. Leaders of the tra ditiona list f a ction r e ­
sntial­
alize that an attempt forcibly to impose sanctions against th e le aders
e they
of th e Gu jax faction would pose a mo r e serious threa t to tribal unity
nswers
than the pote nt i al loss of some unus e d pastureland to th e Guj{rbal)<? e vols.
les
A continuation of t he narrative above illustrates on e way in which the
Gujar faction ope rates . 1-1h.ile the Kom leade rs were still in Kabul \Vord
nviron-­
came to them th a t the leaders of the Guj a r facti o n wer e trying to muster
ed in
suppo:r:i~ for the wali r s pla n. '1'11e led ders in Ka.bul se nt \\'ord back to
prin­
Nuristan say ing that tho se who \Ve re challenging th eir resolve were acting
ons
against tribal un i t y and s hould be f ined . However, th e le aders of the
trevor)
Gujar factio n h ad gathere d e nough backi ng to t hwar t a ny punishme nt. By
a Hithholding their share they ·the n c au s e d th e fundi ng that sustaine d the
es representatives in Kabul to collapse, fo rc ing t h e r ep resentatives to
Suf­
r eturn t o Nurist an.
ions
o
Having returned to Chaghan Sa ray the wali ordered the Kom to reswne n e go­
tiations. Out of frust:t:'a tion and resentment the former lead e rs refused
to continue as tribal r ep res e ntativ e s. Th e l eaders of th e Gujar faction
,es o f
stepped into th e v a cuum and asserted th emselves as new rep resen tative s .
suc­
They concluded an ag reeme nt t .ba t preserved th e status quo: it allowed the
'ibesmen
Guj a rs unmole sted acces s to speci fi ed p3.sturelands (llihich th ey acknO\','ledged
,n
as Korn territory) in return for a set payment an d it r equired fin es f rom
:ion for
any Gujar c~l gh t deviating from th e specified areas.
le Afg ha.n­
:ions on The net re su lt of a ll this litigation was almost nil. Helpe d by pressure
from the wali th e Kom u e re able to collect th e specified grazing fe es for
.r own two years i but the follO\','ing year the Guj ars once a gain r e neg e d claiming
:, th e that ·the pastureland was public domain. Ho s tiliti es were about to r eS WTLe
lloited
but were hal te d by the adv en t o f th e r epubl ican coup d' e tat (of 1973). In
SLlP­
a cycle tha t has continued since th e inception of the conflict with the
:tge. 'rod.ay
Gujars the situation h a d once again gone from hostility t o litigation to
hardly
h os tility, costing the Kom de Clrly in time a nd re sourc e s and reinforcing th e
bitterne ss th at divides the traditionalists from the Gujar faction.
Under the type of acephalous political organization found in Kom society
tribal unity in times of crisis is maintainable only if all citizens are
willing to make equal sacrifices. It violates an individual's sense of
fairness and pride if he sees fellow tribesmen getting away with non­
compliance. If a minority refuses to cooperate the majority lose s its
resolve and tribal unity collapses.

Both the Gu]'irbandevols and the Afghan government exploit this pote ntia] fo!
divisiveness in order to achieve their goals. They operate symbiotically;
the Gujirba~~evols act as a handy wedge which the government us es to sp lit
tribal unity and in so doing the government sanctions the Gujirba~devols'
continued invasion of Kom territory. The effect of this symbiosis has
been to polarize the Kom into a faction that clings to the traditional
view of the Kom as an ethnic group tryinq to maintain its identity in a
hostile interethnic environment versus a. faction that sees the Kom as a
vanquished people whose future course lies in acquiescence and homo­
genization with the rest of the Afghan state. It remains to be seen hm.,r
or whether the Kom will survive this two-pronged threat to their ethnic
integrity.

14

 ~~----

ty Ethnic Relations and Access to Resources

re in Northeast Badakhshan
f
M. Nazif Shahrani
University of Nevada, Reno

.a] for
Llly; As a means of adaptation for individuals and collectivities within the
;plit changing socio-ecological conditions of their environment, ethnicity is
)ls' a dynamic phenomenon, subject to temporal redefinition and reorganization,
with potential for defining structural integrity, distinctiveness and
effectiveness for people so organized. This paper examines the changing
a nature of ethnic identities and intergroup ~elationships in northern
a Afghanistan in general and in the northeastern frontier regions of
Badakhshan in particular - i.e., the Pamir and Wakhan Corridor areas.
10\1
i.c * * *
The predominantly Turkic and Tajik areas to the north of the Hindu Kush
and Koh-·i-Baba mountain ranges were clail11ed by Alunad Shah Durrani (d. 1772)
shortly after the creation of the independent state of Afghanistan in
1747 at Kandahar. Ahmad Shah received help in his bid for these ter­
ritories from another soldier of fortune, an Uzbek officer named Haji
Khan, Ahmad S~ah's one-time comrade-in-arms in the army of the Persian
monarch Nadir Shah Afshar (see NacGregor 1871: 142; and Burns, et al., 1839:
98). Consequently, the annexation was relatively bloodless.

These northern regions which later became known as Afghan Turkistan con­
sisted of a number of principalities of various sizes, each ruled by a
local !~an with nominal allegience to the Pashtun monarch Ahmad Shah
Durrani. Towards the end of the reign of Timur Shah (1772-1793), however,
Quwat Khan, a member of the Qataghan tribe of Uzbeks established himself
in Kunduz and proclaimed complete independence from the Afghan monarchy
(Burns,et aLi 1839:98). This set the stage for almost a century of
local political strife and jockeying for power among Uzbek and Tajik khans
and mirs of Turkistan and Badakhshan. These factional struggles were
based ( as among the Saddozai, Barakzai and other Pashtun tribes in the
south] on the idiom of segmentary opposition of kinship and/or ethnicity
and always characterized by tyranny, intrigue and tragedy (fuller
historical details are discussed in Shahrani, forthcoming).

This paper is based on field research in the Wakhan Corridor and the
Afghan Pamirs between 1972 and 1974 supported by the Foreign Area
Fellowship Program of the Joint Committee of the Social Science Research
Council and the American Council of Learned Societies and by the Wenner­
Gren Foundation for Anthropological Research. The generous help of
both institutions is gratefully acknowledged.

Ed. note: This is a portion of the paper delivered at the MESA meeting,
New York, November 1977. A fuller account of the history will appear
in Soviet Asian Ethnic Frontiers, William McCagg and Brian Silver, eds.,
forthcoming.

15
Generally throughout the eighteenth and nineteenth centuries, no pro­
longed political dominance by any single group over the entire ter­
ritory of northern Afghanistan was achieved. Characteristically, petty
local leaders demanded allegience from leaders of subjugated groups but
made little attempt to assimilate members of minorities politically. or
socially. Copversely, the subjects of these states did not claim any
rights, demand any privileges or have any expectations from their rulers,
and none were extended. During times of peace, the relationship between
various centers and the peripheral areas was more a matter of political
stalemate than of active administrative control by a central authority.
Conflict between various and successive local centers, however, was con­
stant and marked by the rise and fall of local khanates.

The contemporary spatial distribution of ethnic enclaves throughou~

northern Afghanistan and the relative political strengths of the various
groups reflects residues of this history in the nature of their access
to resources and their contrasting statuses in relation to each other.
On the one hand Uzbek and other Turkic speaking groups which were politi­
cally strong inhabit the low lying fertile central valley floors through­
out most of Turkistan. In Badakhshan, likewise, a number of Uzbek tribes
and some Sunni Tajiks jointly occupy some of the more productive valleys
including Kishm, Argu, Darayim, Khash, Jirm and Baharak. Other relatively
fertile but narrow river valleys of the upper Kukcha river and its tribu­
taries are claimed by the SunniTajiks. On the other hand, politically
weak Tajiki-speaking Sunni Hazara are found on the higher reaches of these
central valleys. The Ismailite Wakhi, Ishkashimi, Sanglechi, Kurani,
Munjani and Shighni, all of whom are distinct population categories with­
out political clout, inhabit comparatively marginal and less productive
lands in the upper reaches of the Oxus, the headwaters of the Kukcha and
its tributary Yumgan and Warduj valleys. This pattern of spatial distribu­
tion of ethnic population and dimensions of power, particularly in Badakh­
shan, has changed little since the independent Turkistan period (see
Kushkaki, 1923).1

Ethnic and tribal political processes of 19th century Afghan Turkistan

had two notable characteristics. First, they were not mutually destructive
only internecine • The ethnic name" "Turk" applied to all those who spoke Turl"
or Turk teli (Turkic language) and who were members of one of the following
tribal groups: Uzbek, Turkmen, Kazakh, KtLrghiz. In addition to their own
version of the Turkic language, members of all groups were able to under­
stand and converse in the literary form of Turk tell. They also collecti vel;
identified with Turki speakers to the north and west of the Oxus, as well
as those of Eastern (Chinese) Turkistan. In spite of inter-tribal con­
flicts, Turkic peoples generally rallied as a collective political force
against non-Turkic popula.tions. As a dominant political group they occupied
the most extensive and fertile agricultural lands and pasturage in

ISimilar spatial distributions of ethnic populations in the Bamian valley

and adjacent areas of central Afghanistan are reported by Robert Canfield
(1973a and 1973b). Similar patterns of population distribution and allocat~
of resources prevail in other parts of the country both on macro and micro­
environmental scales.

16

 Turkistan and controlled all the major strategic trade centers and trade
routes of Turkistan. In conflicts among them, defeat, however violent,

ty did not mean ethnic demise or destruction. Turkic dominance remained

ut intact.

["

Second, no single defeat was seen as final. The various khanates of

ers, Turkistan all lacked a centrally organized administrative structure.

een Political influence outside the tribal territ01~ was achieved and main­

al tained either by actual use of military force or the threat of it. Con­

y. sequently, to mitigate loss of life and destruction of property \",hen

on- threatened with military attack, the weaker political community either

retreated to a more distant and less hospitable environment or submitted

to the rule of tyranny and showed allegience by payments of tribute in

the form of goods, money, valuables and slaves. In neither case did it

.ous give up the expectation of a dm"ran (turn to rule and be free) through
,S rebellion. There was a common belief that political power never remains
permanently with any single ethnic group, tribe or family and that all
.iti­ groups or families will one day have an opportunity to exercise their
lUgh­ share of political authority. In other words, both political dominance
:ibes and subservience are transient, a belief succinctly expressed in a 1\.12.ghiz­
_eys Kazakh proverb: "Eluu jilda el jangirat" [A nation regenerates in fifty
:ively years], (quoted in Allworth, 1973:3).
:ibu­
lly These characteristics of the ethnic and tribal conflicts of the past
these affected the ways people of the region adapted to subsequent political

developments.

dth­
Lve
* * -;.,
and In 1884, four years after the beginning of the reign of Amir Abdur Rahman,
;tribu­ all of Afghan Turkistan including Badakhshan was brought under the control
,dakh- of the Kingdom of Kabul. His methods of tribal and ethnic pacification
were not radically different from those practiced in the area before him ­
including among other things deportation of leading households to the
capi tal or to distant territories :.-:--". in some instances summary execution
m to eliminate any potential threat. But he also had an ideology of ­
:ructive creating a politically and culturally unified Afghanistan free of tribalism
loke Turki and "feudalism." With the help of a relatively efficient police and
.lowing administrative organization and a standing army, Amir Abdur Rahman for the
: own first time instituted direct rule of the territories of northern Afghanistan
lder­ under the Kingdom of Afghanistan. During this period, Afghanistan's
_ectively northern borders, including the Wakhan Corridor and the Afghan Pamirs,
well were delineated and recognized by Russia, British India and Afghanistan.
>n- Recognition of these new boundaries marked the beginning of an attempt to
,rce isolate the Turkic populations of the region from the larger Turkic political
)ccupied community of Central Asia across the Oxus and the Pamirs which later led
to the effectual cultural and socio-economic isolation of these peoples.

The peoples of Turkistan accepted the authority of the new Kabul government
,lley without much resistance except for minor revolts in Maimana (1882), Shighnan
:ield and Roshan (1882) and Badakh~han (1889) (see Dupree 1973:419): This lack
lilocation of reaction on the part of Turkistanis and Badakhshis, I believe, was due
micro- to two facts. First, the prevailing Kabul authority put an end to the

17
chronic warfa.re in the area \."hich sapped the human resources of the in­
habitants; and, second, the terms of submission to the alien political
authority were about the same as to the local khans - ~.~., payment of
taxes and a show of allegience. For many small minority e thnic groups,
Pashtun merely replaced Turkic or Tajik sovereignty over them and the
relationship between the local population and the state of Afghanistan
continued on the sa.me basis as \."ith the indigenous Turko-Tajik Khanates
of the earlier period. While the subjects were obliged to pay taxes and
other tribute to the govermnent, they did not have any rights or clairns
on the political authority. The character of this relationship was, indeed,
one of passive submission and not active political support.

This general attitude of inactive participation on the part of the popu­

lations of the northern provinces (then known as Afghan Turkistan) in the
political processes of the country continued through the reigns of Amir
Habibullah, Amir Amanullah, King Nadir Shah and the early part of the
reign of King Zahir Shah. At the same time, during the half century after
the death of Amir Abdur Rahman (1901) the authority of the central govern­
ment itself gre\" stronger.

Considerabl e population changes took place amid these circumstances. First,

there was a significan'c Pashtun incursion. Abdur Rahman relied on Pashtun
support and provid.ed to Pashtun a.mple incentives to settle in the north.
The first large-scale Pa shtun immigration into the northwestern territories
of A£(lhan Turkistan occurred during the 1890s when Amir Abdur Rahma n per­
suaded his political rivals, the Ghi.lzai Pa shtun pas toral nomadic tribesme n,
to move in and occupy the region. By 1910 some Pashtun and Pashtuo-speaking :
luch herde rs had reached the Kunduz areas in central Turkistan. Hore Pashtur
.!!I.aldar (nomadic herder~3) arrived in Turkistan during the 1930s and 1940s
and began taking their herds on long seasonal migrations to the La ke Shiwa
region of Badakhshan and other high pastures on the northern slopes of the
Hindu Kush mountains. The large number of Pashtun nomads displaced some
Uzbek and Tajik comrnuni ties and alienated some of their agricultural and
pasture lands, which had an important impact on the nature of inter-ethnic
relations in the a~-ea (cf.) Tapper 1973, Kaka:c 1971, Dupree 1975).

Furthermore, Pashtun colonies of military and administrative personnel and

their fa.milies had been established prior to the turn of the century in
nearly all ma.jor towns as well as some rural areas in Turkistan, including
Faizabad, the capital city of Badakhshan province. These early official
colonies, \."i th an ever increasing nUll".ber of Pash otun military and civilian
officials who either received land grants from the government or \"ho
bought public land and invited kinsmen and tribal members to join th e m,
later developed into sizable communi ties wi thin towns knovm as Deh Afghanan
or Guzar-i-Afghani, as in the case of Faizabad. In rural areas such set­
tlements were usually referred to by the tribal na.me of the settlers (see
Kushkaki, 1923:174). It should be pointed out that until the early 1950s
all military and police officers and most civilian officials (plus their
entourage) in the northern provinces were drawn exclusively from among
Pashtun or Tajik from the south of the Hindu Kush. Consequently, in addit ic
to the Pashtun colony in Faizabad there is a sizable Pashtun settlement in
Baharak, as well as a smaller one in I shkashim at the entrance to the \<lakha:

18
 Corridor. Both of these areas are located in militarily strategic areas
and have relatively fertile land.

Sizable Turkic (Uzbek, Turkmen, Kazakh and Kirghiz) and Tajik populations
also immigrated into Afghan Turkistan from north of the Oxus during the
1920s and 1930s following the Communist take-over of the Central Asian
s khanates (Dupree 1975: 405). Among them a group of some 2,000 Kirghiz
nd herders left their traditional pasturage territories to take permanent
.s refuge in the Afghan Pamirs. Prior to this flight and the consequent per­
ndeed, manent year round confinement on the "roof of the world," these Kirghiz
had little contact with the people of Badakhshan and the inhabitants of
Wakhan. However, they have reluctantly had to es·tablish relations with a
number of etru1ically distinctive communities under the new conditions in
the Badakhshan. ~~ile circumstances surrounding the Kirghiz entry into Badakh­
r shan were considerably different from those of the Pashtun who came to the
province, both groups had one thing in common: they were both culturally
lfter distinct populations who had not had ext.ensive contact with the resident
'ern- populations of the area and who had to create a niche for themselves within
a new socio-economic and political-ecological environment.

First, . * **
;htun
:h. International and national political developments in Afghanistan have had
:ories a substantial. influence on the processes of adaptation of both Kirghiz and
Pashtun groups in Badakhshan during recent decades. To begin with, the
lesmen, implementation of closed border policies by the Soviet Union and Communist
caking Ba China have effectively ended socio-economic and direct cultural ties that
Pashtun Kirghiz and other Turkic groups enjoyed with the larger political com­
±Os munity in Turkic Central Asia. The Soviet nationalities policy attempted
3hiwa to weaken existing pan-Turkic identity by forging separate "national"
f the identities for each linguistic group as member republics of the Soviet
)me Union. This policy has had a negative impact indirectly upon the Kirghiz
:lnd as well as other Turkic populations in Afghanistan. A significant aspect
thnic of Soviet nationalities policy was the abolition of the use of Turki or
Turk teli, written in Arabic characters, as the common literary form and
medium of instruction in soviet Central Asia. In its place, the use of
1 and different Turkic and non-Turkic languages, ~~ritten in the cyrillic alphabet,
in was instituted. As a result of new language policies first in the Soviet
uding Union and later in Chinese Turkistan, the production of large amounts of
ial material in Turki for readers in Turkic Central Asia by presses in northern
lian India came to a complete halt. The peoples of Afghan Turkistan had de­
pended upon urban centers to the north of the Oxus and the north Indian
m, Turki publications for much of their educational and literary materials.
ghanan The consequence of these developments for the Turkic speaking populations
set­ in northern Afghanistan has been not only a contemporary loss of social
(see contact with the larger Turkic populations of Central Asia but also the
950s severence of contact with the historical heritage of literary Turkic
.eir languages and cultural traditions. Radio broadcasts in a number of dif­
.g ferent Turkic languages from Soviet Central Asia over the past several
addition decades have provided the only means of contact for the peoples of Afghan
~nt in Turkistan with the spoken languages and oral traditions of the peoples
~ Wakhan to the north of the Oxus. 2

2Afghanistan Radio did not broadcast in any of the Turkic languages

spoken in the country until 1972. After a long parliamentary debate a

19
The government of Afghanistan has never formulated anything comparable
to the so-called "Soviet Nationalities Policy." On the contrary, it has
officially de-emphasized the presence of minority groups in the country
and consistently taken measures to undermine larger ethnic and regional
identities and allegiances. For example, the Afghan government dropped
the term Turkistan, replacing it by the phrase manatiq-i-Shamal (northern
regions), and divided the area a number of times for administrative pur­
poses, each time assigning new names to various provinces. Whether done
consciously or unconsciously, this policy has helped to weaken the
larger ethnic and regional identities of the populations in the north.
This unwritten policy, initiated by the Afghan government in its attempt
to create a new modernized nation state, coupled \"i th a lack of tradi­
tional and modern Turkic literature and education in Afghanistan, has
at the present time effectively weakened the traditional collective
identities of "Turkistani" and "Turk" by reducing them to Uzbek, Kirghiz,
Turkmen, Kazakh, Kara Kalpaq, etc. Thus the Kirghiz \"ho settled in the
Afghan Pamirs (as well as other Turkic refugees in other parts of the
country) were faced not only with the disintegration of a larger political
identity, but were also stripped of group privileges in the context of
the new Pashtun dominated state of Afghanistan. In addition, the Kirghiz
of Afghanistan were further affected by their sheer physical isolation
from other Turkic-speaking communities in northern Afghanistan and had
to cope with the extremely marginal environmental conditions in the
hlgh Pamirs.

By contrast, the Pashtun population '<lhich settled in Badakhshan represented

the politically dominant power and, by virtue of their direct association
with it, enjoyed all the privileges and resources the new nation-state
could offer. These included education (military and civilian), access to
public office, cash income, reclaimed government land and a variety of
other strategic resources and services not easily available to members of
other ethnic groups. This practice by the governing power accorded with
the conventional rules of political dynamics in this part of the world:
the powerful have the right to exploit iJ.nd the ';leak have to submit, perish
or flee. Flight, however, is no longer an option since the takeover of
Muslim Turkic Central Asia by the Soviet Union and China.

While the notion of civil and human rights for the subjugated may have
been entertained by individual rulers in Turkestan or Afghanistan, no
such rights existed, even in principle, until the promulgation of the
first Constitution of Afghanistan (nizillmlarrah-ye-Asasi-e-Daulat-~IAliyah
e-Afghanistan) by Arnir Amanullah in 1923. In this document "the General
Rights of the Subjects of Afghnaistan" were spelled out for the first time
in Articles 8 through 24 of the Constitution, and the spirit of the law
proclaimed equality of all citizens of the state (see Poullada, 1973: 277­
289). Later constitutions in 1931 and 1964 also retained quite idealistic
statements to the same '2ffect. However, as Dupree has remarked,"Until
recently, these rights [of Afghan subjects] were more violated than per­
petuated" (1973: 466) .

forty-five minute long program in Uzbek and Turkmen was introduced as part
of minority "national languages programs" of Radio Afghanistan. The progrill!
was a definite success with northern audiences. For the first time it also
created a dialogue on the air between the peoples of Afghan Turkistan and
Soviet Central Asia. Much to the diffimay of everyone in the region the
entire "national languages program" of Radio Afghanistan was inexplicably
abolished in 1974 by the Daoud regime.
20
 ,.----­

as Most rights and services which were granted to the citizens of the
country on the basis of national laws were extended at differential
rates to different ethnic populations residing in different parts of
the country. For example, until the 1950s educational services were
.ern introduced in Badakhshan and other non-Pashtun or non-Tajik areas at an
ir­ extremely slow pace to limited areas. The medium of instruction was
Ine always Persian or Pashtu and, in some cases, Tajiki and Turkic-speaking
children were instructed in Pashtu, a practice which still continues in
some areas of Badakhshan. 3 Most students from the northern provinces
allowed to pursue secondary education in Kabul boarding schools ,were
permitted to enter only vocational schools. Perhaps more significant
was the fact that until about 1958 no students from Badakhshan of any
ethnic origin were admitted to the military school which trained officers
liz, for the Afghan army. This restriction \vas removed only in the late 1950s

e
when the central government had sufficiently strengthened its military base.

:ical Similarly, health care and other social services in Badakhshan were in­
troduced ' slowly compared to other parts of the country. There has been
jhiz virtually no appreciable public investment in any kind of economic develop­
ment anywhere in the province despite the fact that the economy of the
a province suffered considerably vIi th the closure of trade routes to Chinese ·
Turkistan. Many Uzbek and Tajik caravan traders from the province ex­
perienced considerable financial losses as well as the loss of social and
political status as a result of the closure of the borders. Badakhshan
sented province has remained the least developed and the regional economy is in­
tion creasingly drained by a flood of non-essential but expensive consumer
goods from the outside.
s to
if * * *
5 of
ith Improved roads and market demands for raw materials, together with the
d: termination of regional trade to and from Chinese Turkistan created par­
erish ticularly favorable conditions for the influx of traders from trading
of centers in other parts of the country. Most of these entrepreneurs are
Pashtun and Tajik immigrants from areas south of the Hindu Kush. The new­
comers virtually control the truck and bus transportation system throughout
e Badakhshan. In addition, a smali group of Pashtun have dominated the used
clothing market, the tea trade and the only comme rcial export-import com­
pany in Badakhshan.
y ah

rral
time 3 In addition, all school textbooks and popular histories published in
Afghanistan emphasize and often exaggerate the role of Pashtun in the
.aw
277­ development of political events in the region, while the role of Turkic­
istic speaking and other minority groups in the history of the area is frequently
,1 ignored, misrepresented or presented in such a manner as to convey erroneous,
er- negative images of their part in Foli tical processes.: Consequently , despite
alleged equality of Afghan citizens, Afghan school children a,re told that
Afghanistan is primarily the product of Pashtun efforts. The negative
psychological and sociological effects of this intentional or unintentional
; part practice by Afghan educators upon the identity formation of non-Pashtun
program
youth is undoubtedly eno~~ous. Unitl the 19608, for instance, many Turkic­
.t also speaking school children denied their Turkic identity and tried to ' pass

1 and as Tajik whenever possible, a practice encouraged and accepted by school

e officials. Unfortunately, it seems unlikely that the content of school
:ably textbooks il.nd · histories published in Afghanist.an will be corrected.
Unlike the Pashtun nomads and officials, the penetration of Pashtun traders
into Badakhshan has not been limited to market towns or summer pastures.
On the contrary, during the past decade their presence has been felt
everywhere. A group of very enterprising Pashtun itinerant traders has
entered the area of Wakhan and the Afghan Pamirs and their impact on the
local economy as well as inter-ethnic politics has been marked. Pashtun
are not, however, the only outside traders operating in the area. A number
of Uzbek and Tajik itinerant traders from the villages and towns of central
Badakhshan frequented these frontier regions even before the arrival of
their Pashtun competitors. The nature of transactions among the ethnically
diverse traders and Wakhi and Kirqhiz inhabitants of the Corridor under the
current political and economic conditions are of interest for two reasons.
First, they represent new forms of socio-ecological adaptation and inter­
ethnic competition for economic resources, mainly through trade and exchange
rather than armed struggle. Second, they permit an examination of inter­
ethnic relations at the local level under the new conditions and of con­
sequent ethnic claims to differential statuses as these are reflected
through exchange systems among members of different cultural categories.

Kirghiz speak a Turkic language of the same name, Kirghiz. They are
relatively conservative practitioners of the Hanafi school of Sunni Islam.
Kirghiz have inhabited the high valleys of the Afghan Pamirs (altitudes of
from 13,000 to 16,000 feet) for half a century. Despite the loss of much
of their traditional pasture land and of their socio-economic and cultural
ties with other Turkic communities of Central Asia, they have successfully
managed to retain their pastoral nomadic mode of adaptation in the high
Pamirs (see Shahrani, 1976a, 1976b, 1974 and in press) .

Wakhi are mixed farmers and herders in the Upper Oxus valley (altitudes of
9,500 to 11,500 feet). They speak Wakhi, an archaic IndO-Iranian language,
and adhere to an Ismaili sect of Shira Islam. They refer to themselves as
Kheek and their features are Iranian in comparison to the more Nongolian
appearance of the Kirghiz.

Individual itinerant traders who operate in this region are from rural and
urban ,:u:eas outside the Corridor. Some of them have become permanent
residents of vJakhan and have acquired large land holdings in the area. The
majority of the approximately 35 traders are either Tajik or Uzbek speakers
from central areas of Badakhshan. About ten Pashtun traders come from a
village near Jalalabad. The Pashtun all seem to be related by kinship and
marriage; a few of them are in partnership. All of these outside traders
admit to having been less successful in their economic ventures in their
original community and their success in Wakhan and the Pamirs varies
greatly from one to another. All the traders are Sunni Muslims. They have
varying degrees of competence in vernacular Wakhi and Kirghiz.

The traders maintain regular, direct contact with Kirghiz and Wakhi; they
also have firsthand knowledge of wider market demands for agricultural and
pastoral products. Successful traders seem to make full and effective use
of local, regional and national political and economic realities to further
their own interests. They are not only the economic middlemen linking
primary producers with national market economies but also agents of social
change and an important force for the development of the Kirghiz pastoral
nomadic and Wakhi agro-pastoral subsistence systems which ultimately pera
petuate the ethnic identity and separate communities of the Kirghiz and

22
 ------

Wakhi inhabitants of this frontier region.

aders
s.
Itinerant traders organize and perpetuate triadic trade and exchange
relationships involving the Kirghiz and Wakhi in the larger regional and
s national economy. In this process the traders maintain strict control
he
over the supply, type and amount of different market goods and the selec­
un
tion of pastoral and agricultural products acceptable in exchange for
Umber market goods. Traders have also fostered demand, even dependency, among
ntral farmers for market goods and have relied on credit or delayed exchange
f
rather than direct and immediate exchange to maximize their profits.
ally
The choice of items brought into and taken out of the area is influenced
r the most significantly by bulk, weight and the margin of profit to the trader.
ons.
As a result selection of imported market goods disproportionately favors
er­
harmful "luxury" items such as tea and opium.
change
er-
Among the traders the Pashtun have been the major suppliers of these two
n- items, particularly large amounts of tea. In their economic exchange with
Wakhi, Pashtun traders determine prices, terms of credit and, on occasion,
s.
induce Wakhi to buy goods by means of threats or deception. The traders
generally have the cooperation and tacit approval of officials because of
ethnic association, kinship relationships or bribery. While problems
lam.
between Wakhi and Pashtun traders could end in the courts, Wakhi will
,s of
generally meet Pashtun demands. Pashtun superiority over the Wakhi is
~uch further demonstrated by the number of Pashtun who have taken wives from
.ural
;fully Wakhi, bought Wakhi land and settled in the area; the reverse never happens.
Ih
Economic and social interaction between Pashtun traders and the Kirghiz is
somewhat different. With the Kirghiz, Pashtun traders operate on the basis
of uniform rates and terms of credit regardless of the social position of
~s of
the individual Kirghiz or their place of residence. This is generally due
luage,
to the strength of the Kirghiz' kin-based local political organization and
~s as
the absence of Afghan government administrators in the Pamirs. The Kirghiz
.an
khan often negotiates the exchange values of commodities with the traders
and, once settled, the rates are followed by all traders. Disputes are
rarely taken to government officials but are generally resolved through the
and
local political leaders - Khan, Be or Aqsaqal. There have been no exchanges
of women between Kirghiz and Pashtun. and the likelihood of intermarriage
The
seems remote.
~akers
la
Uzbek and Tajik traders operate on low budgets and most of them deal mainly
) and
in trinkets and opium, although they may obtain some tea on credit from
lers
their Pashtun counterparts. Their attitude toward Wakhi is contemptuous but
~ir
in dealing with them they do not generally resort to threats or engage in
deceit. However, different rates and terms of credit are available to
, have
individuals on the basis of social position and rapport with the trader.
Generally their interaction is amiable. Tajik and Uzbek from Badakhshan,
traders and others, have married Wakhi women; but no Uzbek or Tajik women
:hey
have been given to Wakhi men. The Wakhi, together with their neighbors the
Land
Kirghiz, have twice elected an Uzbek trader who lives in Wakhan to the
~ use
Afghan Parliament as their representive during the latter part of the 1960s.
lrther
Uzbek and Tajik traders' economic and social relations with Kirghiz are on
an equal footing. Some traders (especially Uzbek) have established permanent

23

. -

-­
partnerships with individual Kirghiz households and enjoy a great deal of
help and respect. They observe uniform rates of exchange and ~el::r.Is of
credit. Any conflict of interest is resolved through negotiation and
the use of local mediators such as the khan. Violence or resort to the
courts is rare. Both Uzbek and Tajik traders have married Kirghiz women;
and although no women from either group have been given to the Kirghiz,
there are no cultural objections on either side.

Status differences between Uzbek and Tajik are extremely hard to detect in
Wakhan or in other parts of Badakhshan at present. All forms of exchange,
including political support and exchange of women, are carried out with no
reservations on either side. Pashtun, on the other hand, claim a higher
status over both Tajik and Uzbek which is seen in some exchanges; Pashtun
have married both Tajik and Uzbek women but Pashtun women in Badakhshan
married to Tajik or Uzbekmen, although not unheard of, are few.

Perhaps the most significant status differences are observed in exchanges

between Kirghiz and Wakhi. The Kirghiz refer to Wakhi as Sart (a deroga­
tory term) and regard them as non-believers. Feelings of contempt are
mutual yet both groups have developed an increased economic dependence on
one another. The Kirghiz, who cannot produce their own cereals in their
hiqh. altitude habitat, depend on Wakhi for grain, obtained either cirectly
frQm Wakhi or indirectly through traders. The Hakhi, on the other hand,
depend on Kirghiz for anima~s and animal products both for subsistence
and for paying the traders who offer better ex:;nange rates for pastoral
products than for agricultural produce. ~vakhi and Kirghiz, who had very
little contact with each other prior to the closure of the Soviet and
Chinese borders, have had to establish close socio-economic ties with each
other. They have achieved a successful economic exchange system in a
situation filled with social tensions.

Both groups travel freely to each other's territory for trade and they ex­
change a variety of agricultural, pastoral and, at times, market goods.
However, members of each group conduct themselves on these occasions in
ways that communicate attitudes about their status claim vis-a-vis each
other. While economic exchange moves both ways on the basis of market
principles, other forms of exchange are quite asymmetric. For example,
while the ~'Jakhi eat food cooked by the Kirghiz, Kirghiz rarely eat with
Wakhi. Kirghiz often spend month~ during the winter in Wakhi territory on
trading trips spending most of the time in Wakhi households. Yet Kirghiz
eat nothing cooked by Wakhi except tea. Kirghiz hire both Wakhi men and
women to perform menial tasks for them but a Kirghiz will never be found
working for a Wakhi.

Conflicts between the two groups are rarely, if ever, taken to the courts
staffed by Pashtun and others from outside the Corridor. Instead they are
resolved through negotiation or by Kirghiz threats of aggression. I have,
however, encountered situations where the Wakhi have been accused of ini­
tiating aggression against individual Kirghiz, generally in \'Jakhi territory ,
Exchange of women or even the suggestion of sexual relations with Wakh;
women outrages Kirghiz males; giving a Kirghiz woman to a Wakhi is
unthinkable.

24
 ~~----

.1 of Perhaps the most vivid symbolic expression of the sharp value contrast
If Kirghiz see between themselves and their neighbors is demonstrated in an
t episode which I recorded during my field work. An old Kirghiz man died
:he while in Wakhan on a trading journey in the winter of 1973. Such a
omen; situation had not arisen before. His kinsmen and companions refused to
.z, bury him in Wakhan, "the territory of the non-believers." Instead, they
transported the corpse on horseback to the Pamirs, a journey of four days,
so the man could be properly buried in Muslim soil.
~ct in
lange, *' * *
.th no
Jher On the basis of this discussion, a number of points may be emphasized.
lshtun First, the dynamics of local political processes as well as social and
lan economic intercourse in northeastern Afghanistan historically have been
dominated by ethnic and tribal conflicts and competition for power, privi­
lege and access to strategic resources. Second, allocation of social
mges services and economic development projects are, at present, governed
t oga­ by an idiom of kinship as well as by ethnicity and spatial distance of
[e the periphery from the center. Third, the traditional petty states of
ce on Turkistan, as well as the early Afghan monarchies, operated on prin­
~eir ciples of exploitation of subjects by rulers where subjects had no rights
rectly and could make no demands on the state. Reaction, or expression of dis­
:md, content, was by means of retreat or revolt whenever possible. These
ce options, however, became impossible vis--';;'-vis the modern Afghan state due
ral to its increasingly strong military base, created with the help of foreign
very governments, and the prevailing condition of closed borders. Therefore,
d for a long time the traditional outlet for ethnic or tribal discontent
h each has been absent but no alternative mode of expression has yet developed.
a Lastly, the submission of the Turkic and other minority groups to the
rule of dominant Pashtun authority has been realized and the larger
ethnic and regional identities of Turk and Turkistani effectively weakened.
ey ex­ With the increasing spread of education in all parts of the country, how­
ds. ever, attitudes are changin~ and the expression of demands for rights
in and privileges along ethnic and class lines as in the liTest may, of course,
ach come.
et
Ie,
ith
oryon
rghiz
! and

ound

:ourts
ley are
have,
. ini­
!rri tory.
fakhi

25
The Impact of Pashtun Immigration on Nomadic Pastoralism in Northeastern
Afghanistan

Thomas J. Barfield
Harvard University

The Pashtun emigration to Qataghan province was the last of the great
migrations that put large numbers of Pashtun settlers into northern
Afghanistan. Pashtunization of northern Afghanistan, the home of
Turkic and Tajik peoples, had been a goal of the Afghan government since
the time of Amir Abdur Rahman (1880-1901). This region was particularly
attractive because it was potentially the richest part of the country,
and because it was the frontier with Russia where politically reliable
Pash -tun provided security for the Pashtun government in Kabul. The
large number of immigrants involved, the importance of government aid
and the success in Pashtunizing Qataghan provides a case history of
Pashbm penetration in an Uzbek province where interethnic competition
was regulated by state policy. Specifically, I examine the recent ex­
pansion of Pashtun pastoralists into northern Qataghan and the dynamics
of their competition and eventual coexistence with the Arab pastoralists
already established there.

Qataghan in the 19th Century. Northeastern Afghanistan ,-!as traditionally

divided between the lowland river valleys of Qataghan and the highland
mountains of Badakhshan. Geographically Qataghan encompassed the Kunduz,
Khanabad and Amu river valleys as these rivers left the mountains for
the loess plains. There they formed large swamps inhabited mostly by
Uzbek farmers through the 19th and into the present century. The name of
the province was in fact taken from the dominant Qataghan clan of Uzbek
who ruled the region. Because of endemic malaria, Qataghan was com­
paratively sparsely populated.

The Uzbek in the valleys were both farmers of irrj_gated land and sheep
raising semi-nomads. The land was fertile and produced large surpluses.
A British report of the 1830s praised its productivity: "As for grain its
production in this country is limited by its being all but unsaleable. Any
man who chooses may have ground to cultivate on the condition of paying an
eighth to the Arueer ... There is probably no country on earth where life
can be supported cheaper or better. Though money is scarce there is no
absolute poverty." (Burnes et al., 1839: 131).

Those Uzbek not engaged in full time irrigation agriculture used the valley
for permanent winter villages (gishlog) and moved to the grassy steppes
and foothills above the valley floor to graze their sheep during the spring
and summer. These migrations were short and were often combined with some
form of sedentary agriCulture.

Research among Central Asian Ar"abs in Afghanistan between January 1975

and September 1976 was supported by a Fulbright-Hays Dissertation
Research Grant.

26
 ~---

~rn In the foothills themselves some Uzbek, or closely related Turkic people
like the Moghal (Schurman J962: 99;101), engaged in whe at farming or non­
irrigated fields (lalmi). These villages marked the limit of Uzbek set­
tlement of the plains and river valleys. The lalmi farmers v,ere more
nomadic than lowland valley Uzbek. For example, Moghal in Argu took their
sheep for the summer to the high mountains of Shiwa in Badakhshan and Uzbek
from Rustaq took sheep to Darwaz (Shurman 1962:100, G.A.B. 1972: 59), but
both were economically dependent on agr iculture.

In mountain valleys and hilltops above the Uzbek lowlands were the Tajik
of Badakhshan. They were f a rmers of lalmi land and tran shuman t herders
of cattle and goats. Depe nding on elevation Tajik grew wheat or barley,
nce the highest villages being entirely aependcnt on b a rley. In comparison
rly to the f er tility of the 10\vland it was a stingy land. In many villages
groves of mulberry tr ees provided an important addition ·to grain. Dried
e mulberries could be stored indef initely and ground into flour should the
h a rvest be poor. Villagers took their cattle and goats to luxuriant summer
pastures (ailoq) which were wi thin a day's walk of the village. l-10vement
betwe en the 10vler village and the highland pas.ture was therefore easy.
n Despite large tracts of available pasture Tajik husbandry was severely
limited by the need to stable a nd stall ··fe ed th e ir animals through the
cs harsh mountain winter. Lack of fodder limited the number of animals a
sts Taj ik could keep to the number he could support througl~ tr.e winter.

:Juring the 19th century the Tajik of Badakhshan were conquered by the Uzbek
nally of Qataghan. The re s ult was disastrous. Badal~hshan vlas devastated in an

d atte!Jlpt by Hir iviorad Beg of Kunduz ·to settle the fertile but- deadly ma­
duz, larial swamps:

since the year 1 830 , Badakhshan and the countries subject to,
e of or rather 'chuppowed ' [raided] by MQrad Beg on the northern
ek bank of the Oxus, have been depopulated to stock the plains
of Kunduz and Hazrat Imam. The aggregate of foreigners thus
forcibly transplanted in these unhealthy marshes from that year
to the present time is estimated by the Uzbeks at 25,000
p families, or in round numbers 100,000 ' souls; and I question
es. whether 6,000 of these were alive in - i838 so great had been
its
the mortality in the space of 8 years. (Wood 1872:258).
Any

ng an Morad Beg felt that "because he lives in [Kunduz], he sees no r eason why

fe the peoples of the hills should not live there also. I ventur e d to sug­
no gest that a reason might be that they invariably died ... " (Burnes~t....: .Sll.,
1839: 121). Th e potential of the swamps of Qataghan was obviously limited.

valley Central Asian Arabs v!ere the third major group in Qataghan in the late 19th
·es century. They were pastoral nomads and, unlike the Uzbek, they made long
spring range migrations that permitted the m to exploit the swamps in the winter
I some
when malaria was not a problem and use the extensive high pastures of
Badakhshan during the summer. The Central Asian Arab of Qataghan tradi­
tionally lived in Baghlan (Burn es ~ 2.:1'1 1839: 32), but they were only a
small group until the 1870s. At that time large numbers of Arab fleeing
'5 the Russian conquest of the Zerafshan valley came to Qataghan and Afghan
Turk§!stan. By the 1880s they wer e the second most populous ethnic group
on the Turk ~stan Plain numbering 15,000 households (Kakar'19il: 139). In
the ~lanat e of Bl~hara the Arab had raised huge fat-tail ed sheep. Operating

27
from fixed villages they made a long-range seasonal migration to the
mountains. They sold their sheep profitably in Bukhara and 1'le re closely
tied into the urban markets of the region (Khanikoff 1845: 72-73, 204-205),
Upon their arrival in Qataghan Arab took advantage of the spars e ly popu­
lated swamp and steppe l a nd to establish winter qua:ct e :cs where they
pleased. Arab were fully nomadic, changing \-linter quarters annually
until 1921 when they receive d governme nt land grants in Kunduz and Imam
Sahel, - the latter in the Amu Rive r valley. From that time Arab nomadism
has been b a sed on fix e d \-,'int er villages ( q ~0}og) I,hich were s e asonally
abandoned ~.n the spring to move to the sur rounding s -teppe and in the sum­
mer for the mountain pas -t-ll.res of Badakhshan 300 kilome ters away. Hountain
pasture was plentiful, in part because of the forced mi g ration policy of
Horad Beg that has so drastically reCluced Ba dakhshan's population. Arab in
Kunduz obtained ~iloq in the Sh.h·la dis ·trict of Badakhsha n, tho se from Imam
Saheb took theirs in Da:C\vaz.

The ]I.rab 1<1ho came to Qat.aghan W2re bilingual in Persian and Uzbeki and had
develope d clos e ties with t.he Uzb ek in the Khanate of Bukharai none spoke
Arabic. Socially a_no' poli t.ically the i\rab h a d allied themselves with the
Turkic p e ople against t.he Tajik and maintaine d t.hi s relationship \vith the
Uzbek in Qataghan aft e r their arrival in Z\fghanistan. Even todayL:he Arab
show a marke d dislike of the Tajik. While th e y marry Tajik wome n, they
refuse to marry Arab women ·to Tetj ik men.

At. the end of the 19th century we find e,ach of the thr e e ethnic groups,
Uzbek, Tajik and Arab, holding a p a rticular jliche in a large regional
system. UIban centers and irrigated river va l leys were under the control
of Uzbel~. 'I;urkic semi ~ 'nomadism V: clS common b'Llt involved only short migrat.ion;
Poli tical and rnili tary pOi'Ter had allo-ded the Uzb e k to control the most
fertile vall e ys and plains as well as those accessibl e rl10untain valley
terri tories like Rus 'l:aq, Argu and Faizabad. Lm·,land Uzbek settlement was
Ij,mited by endemic malaria in the swampy vall e ys so that -the popula·tion
density wa s Im. T
•

The Ta.j j.ks, able t.O maintain their indep e ndenc e in the mouutains of Badakh­
shan, had a mixed subsist.ence base of animal husbandry and lal!!-,i agricul­
ture. The 'i'ajik niche is actually more complex than described since the
mountains themselves contained Sunni Tajiks in t.he lower mountain valleys,l­
Rogh, Yafta.l, Shar-i-Bozorg, Jurm, e-tc., and Shiite Tajik, oft e n non-Per­
sian speake rs, in the highland valleys of Darwaz, Roshan, Shoghan and the
Wakhan. This provided an extra dimension of religious different.a"cion I"it.h­
in the major niche of mixed mountain agricultllre and pastoral transhumance
(compar e Canfield 1973a).

The Arab migrate 300 kilomete:cs from their winter t.o sUl11mer pastures to
exploi t both highland and lowland resources. 'I'hey avoide d the malarial
summers of the lowland river valley,s by moving to the steppes and mountains.
In the winter they returned and could use the vast slA'amps as winter pasture.
During the sununer the extensive grasslands of Badakhshan allowed the
pastura.ge of far more animals than the Ta.jik could support, so there was
room for many nomadic group s . Politically, by allying themselves with the
Uzbek, the l-\rab were able to se"ctle cluietly and easily into Qataghan. The
major Arab advantage was that long range migr a tion allowed them to use the
best pastures in each region on a seasonal basis and thus keep the maximum
number of sheep.

28
 Pashtun Immigration. Pashtun control of northern Afghanistan was only
ily nominal until Amir Abdur Rahman solidified Pashtun control throughout
-205) Afghanistan. He encouraged, sometimes forcibly, the movement of Pashtun
u-- from south of the Hindu Kush into the Turkestan Plain betvleen Balkh and
Maimana (Tapper 1973). But the Pashtunization of Qataghan proceeded at a
slow rate because of the infamous malarial,swamps. Potential settlers,
m remembering the fate of the transplanted Tajik and the proverb, "If you
ism want to die, go to Kunduz," showed a marked reluctance to move there. A
tI
Pashtun told me that at one time when a man planned to go from the south
fttm-ain to Kunduz his relatives would hold his funeral before he left, so certain
was his fate.

Fa!in There was one exception to the general Pashtun fear of Qataghan: around the
Imam turn of the century Pashtun nomads began appearing in numbers in south
Qataghan. The Pashtun nomads, like the Arab, discovered that a migratory
life made the region more attractive and less dangerous. They settled in
had
fixed villages in the swamps which they abandoned each spring and surrrrner
oke for the mountains. This pattern was common before the Pashtun came to

the Qa,taghan - of 5,000 households who lived in Khanabad, 4,000 were considered
the nomadic (C~A.M. 1894: 412). Traditional Pashtun nomadism was thus pre­
Arab adapted to the conditions in Qataghan.

Pashtun nomads originally settled in the Baghlan and Ghori districts of

Qataghan. By 1914 'they numbered more than 2,300 families with more arriving
:, each year (G.A.B. 1972:6). A survey taken during the reorganization of the
provincial government in 1921 by then -"lar Hinister Nadir Shah showed Pashtun
;rol nomads (lc>cally referred to as Kandahari, regardless of origin) owned over
rrations 376,000 sheep (Kuskaki-1923:102). By the late 1920s Pashtun nomads were in
evidence along the lower course of the Kunduz River. The majority of them
had sununer pasture in the Khawak Pass area of the Hindu Kush, 50-100
T...'-las kilometers avlay, while the remainder made a much longer migration, 200­
m 300 kilometers to Shiwa in Badakhshan (Schurman 1962: 405-408).

This influx of Pashtun nomads was confined primarily to southern Qataghan.

tdakb­ Around 1910 some Kandahari moved into the Amu Valley of north Qataghan and
~ul­ some Pashtu-speaking Baluch arrived from the west (Tapper 1973:72). But
:he informants agreed that the Pashtun were a small minority in north Qataghan
Leys t lik until after the Second Horld War. until that time the Arab and Uzbek were
)er- locally dominant and stories are still told of how many Pashtun immigrants
the
were forced to flee to the Soviet union to escape Uzbek raids.
,vith­

nance The Development of South Qataghan. Pashtun presence in Qataghan might

have remained limited to government officials and nomads had there not been
a striking change in the ecology and economy of Qataghan itself. But in
to 1933 the most successful development project in recent Afghan history
3-1 began as the Spinzar Company acti vi ties created from the disease'-ridden
'ltains. swamps the richest province of Afghanistan. Three elements combined to
3-sture. make this possible: a strong local governor, investment capital and better
transport. Shir Khan, the new governor, in cooperation with the first
:is
capitalists of Afghanistan used corvee labor to' drain the swamps and reclaim th e
h t.he land. Qataghan quickly became the country's leading producer of rice and
ThE'
cotton. Investment in processing factories for ginning raw cotton and
e the cotton mills gave the province an industrial flavor with planned towns

ximum dominated by the monopoly cotton company - Spinzar (White Gold).

29

. . --
A motorable road through the Hindu Kush via the Shibar Pass gave Qataghan
a strong economic link to Kabul for the first time. Acc e ss to Sovi e t ports
on th e Amu Da rya made cotton export a fairly simpl e matter. Forced sale
of governme nt swamp land to weal thy Kabuli c reated an ince ntive to
development (Dupre e 1973; 437), but the r e al credit goe s to th e local in­
habi tants and n e w immigrants '",ho did the work. Attracted by che ap land
and the d e cre a sed dangGr of malaria. as the s\vamps rece d e d, large numbe rs
of p e ople came to s e ttle in Qataghan, not a bly Pashtun f a rmers from the
south and Turkic refug Ges from the Soviet Unio n. t>1a.laria itself was
eradicated with the h e lp o f the Unite d Nations in the early 1950s (Franck
1955: 28-29). By 1965 th e popul a tion of the Imlland valleys had tripl e d
while the popUlation in the mountains remained static. .Re fl e cting this,
the ethnic composition of valleys c h a ng e d r a dically b:'it Ba dakhsh a n l-emained
as it wa s in the 1 9 th century (Gro t.zbach 197 2 : 74-84). In the valle ys the
Pashtun we re now strongly represent e d. Abdur Rahman's plan for a Pas !1tun
presen c e in the nor thern plains had been accomplished. In the mid-1960s
the breakdown was as follO'."s:

Pe rcentage o f -the Po~lation

Pashtun Tajik Uzbek Turkoman Hazara

Baghlan area 56 30 10 2

Pul-i-~jlumri area 61 14 14 5 6

Kunduz area 41 43 13 2 1

Khanabad area 45 36 6 3 10

Taloqan area 10 10 60 1 10

(Source: Entienne 1972: 83)

The most striking pattern in this chart is that the percenta g e of Pashtun
is high es t in the area where developme nt occurred latest and was mo s t ex­
t e nsive. Thus Taloqan, \.;hich \.;as not p a rt of the reclamation project, kept
its Uzb e k ma jority. Kunduz and Khan a bad, d e veloped in the 1930s, show a
substantial Pashtun population but not a majority, r e fl e cting the freer
policy of land sales and distribution that marked -tIl " i n i t .ial development
of the region. Baghlan and Pul-i-Khurnri, d e veloped in the 1940s, show
Pashtun majorities, reflecting the almost exclusively Pa s htun nature of
settlement the re. This is particularly true for Pul-i-Khumri \"hich was
created as a n e w city in a sparsely inhabite d area. Except in Taloqan the
Uzbek became a s mall minority in the province they once dominated.

With the help of the national government the Pashtun strate gy was to over­
,,,helm the Uzbek with she er numbers of settlers. The Pashtun ,·:ere not as
a rule trying to displace the Uzbek because land was at the time an e x­
panding resource. Pashtun success was due to the position of Qa.taghan as
an underpopulated frontier with pl e ntiful land and WCt2~. Pashtun settle­
ment could therefore proceed around the Uzbek and avoid disposse s sing them.
Conflict with the Uzb e k was n e vertheless in e vitabl e and required government
intervention in def e nse of the settlement policy (Akhra.'Uovich 1962:40).
 ------

aghan With their numbers and government support the Pashtun succeeded in estab­
t ports lishing themselves and quietly but effectively took control of the
sale richest part of Qataghan. By the mid·-1950s the frontier was closing. Sub­
stantial immigration had ended and land prices began to rise reflecting the
end of free land, although prices still remained very lav.' in comparison to
fl in­
land prices in mountain valleys (Grotzbach 1972: 267-78).
la nd
!mbers
he The Pashtun Hove into North Qataghan. Pashtun immigration in northern
Qataghan did not begin until after World War II. Like Taloqan, Imam Saheb
ranck in the Nnu River valley was long settled by Uzbek. At the margins of the
pled main canal system Arab and Turkoman had established qishloq in the 1920s
!his, and 1930s. Given the geography of the va.lley there was no easy way -to ex­
-emained pand the area of cultivable land beyond ,-;hat had been done already by the
ys the non-Pashtun inhabitants without major a_ddi tions to the canal system. Instead
.s!1tun the government chose to irrigate a loess plain, the Dasht-i-Archi, and
960s create a whole new settlement upstream from Imam Saheb. Diverting water
f:com the Kokcha River in 1947, the project created an area of irrigated
land equal to that under irrigation in Imam Saheb itself. This land was
settled exclusively by Pashtun many of whom were exiled there for their
partin the Safi rebellion. Archi is one of the few bazaars in Qataghan
where Pashtu is commonly spoken; elsewhere the lingua franca is Persian.
This preferential land distribution to the Pashtun sudde nly established
their presence.in north Qataghan. Since the land was newly irrigated and
not part of Ir;).am Sa_heb itself, the Pashtun were not in direct cor,lpeti tion
with the dzbek and to a certain extent the status quo was preserved. But
many of the Pashtun in Archi were pastorali .s ts and these people carne into
competition with the Arab. They presented particularly formidable competi­
tion because the Pashtun nomadic pa.sto:r:alists filled the same niche as the
Arab. For the first time the Arab no longer had a monopoly on long distance
migration in north Qataghan.

The Pashtun nomads quickly adopted many as}Jects of I'_=2b pastoralism. 'rhey
got rid of the scrm"ny sheep commonly raised south of the Hindu Kush and
invested in large fat-tailed and karakul sheep. l'Ji th fixed bases in Archi
and dependable pasture the Pashtun found sheep raising much more profitable
lshtun
and reI i ttble than it had been south of the Hindu Kush. Here they were able
i t ex-'
~t, kept to combine sedentary villages with pastoralism. The Pashtun became the
lOW a
majority nomads in north Qataghan as they had in south Qataghan. One
old Pashtun nomad 1 ,·rho came as a boy -to Imam Saheb 60 years ago, told me:
~eer
"I remember '-lhen t..'1e Arab were a big people and the Kandahari only small.
Jpment
Now it is the Kandahari who are the big people and the Arab who are small."
lOW
The change occurred rapidly, without violence, and because of the peculiar
~ of
was nature of pastoralism in Qataghan the Pashtun did not displace but aug­
mented the Arab in the sheep business.
1an the
Pashtun nomads had a number of advantages over the Arab in north Qataghan:
J over­
1) Pashtun nomads traditionally engaged in long distance migration and were
considered, even by the Arab, to be superior at the business of nomadic
Jt as
pastoralism.
ex­
2) Pashtun quickly outnumbered the Arab because of the large new settlement
lan as
project on the Dasht-i-Archi.
settle­
3) Pashtun were better armed and more agg_r essive than the Arab. By cul­
og them.
tivating a fierce reputation they had far less trouble than the Arab in
ilernrnent
dealing with other ethnic groups, especially during the migration.
40) .

31
4) As Pashtun in a non-Pashtun region of Afghanistan they could count on
government aid in disputes, or at least biased decisions in their favor.
It is ironic that often the staunchest supporters of the central govern­
ment in Qataghan "fere Pashtull exiled north for rebellion against it.

Barth (1956) discussed Pashtun expansion in SVlat in terms of niches in

~vhich the limit of Pashtun expansion 'vas the fertile valleys that could be
double-cropped. No two groups could hold the same niche, the competitive
exclusion principle positing that the more efficient would drive out the
less efficient. \17hile this may have been the case in S'-lat, Vlhere Pashtun
dominance was established by conquest, the comp'3titive exclusion principle
",as not operative in Qataghan because peaceful expu.nsion under the
authority of a nation state limited the degree of conflict. In Qataghan
both 1I.rab and Pashtun ,.;ere encapsulated by a higher political authority
that 'vrote the rules of the g·alne and defined what competition 'vas permitted
or prohibited. In this situation politics and government policy backed by
mili tary pm·,er \-lEore as much a part of the environment as the seasonal
grasslands or rainfall.' Both the Arab a.nd Pashtun had to adapt t:o a politi·
cal reality ~lhich maintain e d a pec'clliar but distinctive feature of pastoro­
alism in Qataghan: privately owned spring and summer pastures.

Pasture land is not state, tribal or lineage property but is owned and
inheri ted in single families. It '°/as Nadir Shah in 1921 who created this
situation by goiving rights to Arab who first cla.imed the pasture. His
firman gave them legal title as exclusive us e rs of particular pastures.
This was extreme.ly important because nomads did not wander in Qataghan.
Both the steppe in the spring and the summer pasture in Badakhshan were so
rich and dependable 'chat once nomads moved to their new pasture they stayed
put for t:hree months.

Had all the pasture been initially mmed by Arab trouble would have been
inevitable, but the government traditionally reserved the best pastures
for its ovm livestock. The high plateau of Dasht-i-Ish in Darwaz was
used exclusively for governInent.. herds. As state pasture (Sarkari) it was
prohibi ted to the Arab who had ailoq on the mountain slopes above and belo.
Dasht-i-Ish. The governrrlc nt eventually stoppe d keeping animals, most of
which disappeared during the Saqqaoist uprising in 1929, and 'vhen the
Pashtun arrived iJ;l. Archi they not only received irrigated land but the lar~
pastures of Ish. Arab and Pash'tun nmv both had their own areas with the
same kind of firman. The Pashtun in the steppe were given pa.sture between
Arab holdings and the steppe began to resemble a checkerboard of Pashtun ar
Arab households. Only the swamps were common to all but even these today
are regulated, though illegally, by the border commissar who demands a fee
for each herd in his jurisdiction.

Each nomad family that acquir e d propert:,' rights to a pasture found it

highly advantageous to maintain the status quo. Thus Pashtun uho received
title to pasture felt no obligation tovlard, newly arrived Pashtun to
share pasture or to help them obtain it from Arab holdings. From the very
beginning solidarity among pasture owners was more important than ethnic
ties. Competition for scarce resources therefore was at a family level
and rarely threatened the interests of Arab or Pashtun pastoralists as a
whole.

32
ount on
At present the most critical resource is an ailoq in the mountains for sum­
avor.
mer pasture. Nomads who do not own pasture must obtain pasture from those
vern­
who do if they wish to remain nomadic pastoralists. This can be obtained
in three ways: purchase, rental or theft~
in
Pastoralism is a risky business and because of various disasters a nomad
:ould be
may lose all his sheep. If he owns a good 1,000 sheep ailoq in a sec~~e
titive
area he may decide to sell it. For exampl e , the ailoq in Darwaz are large
It the
and whole areas are securely controlled by Arab. In the mid-1950s an Arab
la shtun lost his sheep and sold his ailoq to a Pashtun for 80,000 afghani. The
'inciple
Pashtun used the pasture but was troubled by Arab sheep thieves, relatives
of the original owner, \vho resented his presence. After ten years of this
.a.ghan
he agreed to pay the former owner an annual bribe of 4,000 afghani to
)rity
call off his relatives. The harassment stopped and relations"became more
~ ermi tt.ed
friendly. Buying an ailoq is a major investment and those Pashtun who have
leked by
purchased them have usually "traded up" from a less satisfactory ailoq and
lal
are therefore already established and fairly wealthy.
a poli ti­
pastor·­
For the poor nomad the theft of an ailoq is a way of b e coming established.
A Pashtun cannot for-ce an Arab out since violence is prol.·dbi ted by the
government; but should an Arab in a marginal area be unabl e to use his
=d and ailoq, a Pashtun will steal it. This does not happen with large ailog in
=d this
Da~7az because other Arab would not permit the squatter to move in and,
His
besides, wealthy Pashtun are more than willing to buy a first class ailoq .
.ll~ e s .
It is the more marginal pasture areas, in the highlands of Rogh, for ex­
jha n.
ample, that are at risk. The ailoq are smaller and owned by poorer Arab.
were so
Should they los e their sheep they can expect no compensation. Since this
?y stayed
interaction takes place between poor Arab and pastureless Pashtun, the
conflict is of little interest to the better establishe d nomads. In fact
many of the ailoq in Upper Rogh were voluntarily abandoned by rich Arab
e been
who owned two ailoq and found the poorer one not worth keeping.
tures
\vas Renting pasture is by far the most common means of acquiring at least tem­
it was proary pasturage. Many owners of ailoq have more pasture than sheep. They
a nd below are willing to rent the excess to other pastoralists. Rentals range from
ost of 4, 000 to 15, 000 afs. depending on the size and quali ty of the pasture. One
t he Arab who owns a 3,'600 she ep ailoq rents it out in 800 sheep parcels and
the larg e
uses part for himself. Contracts are not automatically renewable and
th the price gouging i s not unknO'.-.'n. An Arab who rented a pasture to a Pashtun
between at an inflated price e xplained, "hlhat can he do? He is majbm: [·::;ompelled],
a shtun and he has no ail~'l'" The renting of pasture allows the maximum use of pasture
e today without disturbing the nature of ownership. The system is kept intact
ds a fee ul timately by the government. Until recently troops \vere posted in Dasht-i­
Ish and Dasht-i-Shiwa to keep the peace. l-fuile the government favors the
Pas htun it also grants rights to the Arab, and the government wants
l it stability. It is willing to allow marginal Pashtun to replace marginal
received Arab but ultimately the guarantee of land rights creates a peaceful atmos­
to phere in an area remote from centers of population.
the very
ethnic When competition for critical resources is controlled by a nation state via
level pri vate property, etJ-~ic differences are ofte n less important than economic
;5 as a
differences. As the price of sheep rose in the past ten years wealthy
Pashtun and Arab pastoralists began to have more in common with each other
than with their poorer relations. Private property transferred the burden
of direct competition to the more marginal members of each group. The
33
conservative interests of large pasture owners created a barrier that
allowed Arab and Pashtun to co-exist because access to critical re­
sources was ultimately guaranteed by the central government. Pashtun
had a competi~ive advantage over the Arab, but the state defined the
are,la of competition so that households rather than corporate tribal
groups were the competi t~.ve units. By encapsulating tribal groups the
central government effectively reduced their independence of action.
Pashtun in the north are as effectively dominated by the Afghan state
as any other ethnic group. The Afghan government used Pashtun im­
migration as a tool to gain better control of non-Pashtun regions and
the Pashtunization of Qataghan was a success. But while Pashtun were
the major beneficiaries ·the interests of the government and the tribal
Pashtun were not the same. The government, thro~gh private property
and government protec ·tion of other ethnic groups I rights I came to
control the troublesome Pashtun more effectively than had ever been
possible in their traditional homeland south of the Hindu Kush.

34
 Religious Myth as Ethnic Boundary

n
Robert L. Canfield
Washington University, St. Louis
he

,e

We were convened in this symposium to discuss "evolving interfaces" of

the various minority ethnic groups in Afghanistan with "dominant and
intrusive Pakhtun." In this paper I \\7ant to address the topic by
questioning an assumption implicit in this choice of words. They assume
that the important social distinctions in traditional Afghanistan society
are ethnic. I would like to propose another way to think about that, or
at least to suggest that this assumption ought not always to be made in
discussions of trad~tional socio-political relations in Afghanistan.
It needs rather to be examined.

I shall raise my question by discussing a topic that at first may seem

irrelevant, the s e ctarian myths that are told among the people of Afghan­
istan. By sectarian I mean things associated with the beliefs of the
three Muslim mazhctb in Afghanistan - Sunnis, Imami Shiites and Ismaili
Shiites. ,T he teDJIS se_ct and sectarian may grate on the sensibilities of
some persons but I have no better terms for now. The word mazhab, which
the Afghans most often use for the distinctions I call sects, properly
means "school" - that :;'s, a school of Islamic jurisprudence. The term
designates a tradition of legal interpretation and, therefore, does not
necessarily suggest an organized social c211ectivity such as the word sect
implies. f)ut in Afghanistan, mazhab suggests groupings of people who have
followed different schools of Islamic interpretation. In times past, these
groups have warred against each other; they have organized against each
other. The term mazhab in the Afghan setting connotes more -than mere dog­
matic entities. It suggests socially and politically significant groupings
of people. So the term sec:t seems to me an adequate if not wholly satis­
factory translation for the proper term mazhab as used for an important
social distinction in Afghanistan society (compare Gulick 1976: 168 !.).

The field research on which this paper is based was funded by the Foreign
Area Fello\\7ship Program. A Fellowship from the National Endowment for the
Humanities, a Summer Award by the Center for Near Eastern and North African
Studies at the University of Michigan and several Faculty Research Grants
from the Graduate Schools of Washington University have supported my sub­
sequent research. I am grateful to all these agencies.
I have benefited from comments made on an earlier draft of the paper by Jon
Anderson and Dale Eickelman. Although Karen Blu had not read the previous
draft of the paper, her comments on ethnicity emboldened me to make my point
more forcefully. I also thank Vernon Kramme and Richard Blocher for making
special arrangements for me to prepare a previous draft of this manuscript
through the Computing Facilities of Washington University. I am, finally,
grateful to the editors of this work for allowing me to remove a major
portion of the paper delivered at the orig inal symposium and to develop it
in a different way for another publication.

35
Bt myth I mean stories that the Afghan people tell to express their
conception of the critical social distinctions in their society. Leach
used the tenn "myth" in this sense. Stories such as I ",ill tell here are,
to Leach, told "as a ritual act which justifies the particular attitude
adopted by the teller at the moment of tl,e telling" (Leach 1954: 77). For
Leach "sacred tales have no special cha.'. acteristics which r.nke them any
different from th'2 t a les about local happenings bventy yei:u~s ago. Both
kinds have the same function" (ibid.! 277). They identify the story teller's
relationship to one of several rival social groupings in his society and
indicate his attitude on the issues that rend these groupilJ]s apart (ibid.,
277). They are also myths in a sense that L~; vi-Strauss has expressed :
that myths represent (like everything else cultural for L ~ vi-Strauss) binary
distinc'cions inher en t in huma n thought, expressed in t_he vehicles of thought.
LEivi-Strauss sees myth as a kind of grand metaphor which reveals "a primary
form of discursive thought " (1963: 102) . He believes "the demands to ,·"hich
[myth] responds and th e 'way in which it t r ies to meet them are primarily of
an intellectual kind ' (1963; 104) - or that ilyte llectual understanding operCl.te'
"by means of binary oppositions" and coincides 'vi th "th e first manifestati
of symbolism" (1963:102). I do not knOlv whether this study of Afghan myth
would pleas e Levi- Strauss or other structural anthropologists ; but it
seems interesting to me that, in fact, the Tnythical system to which -these
stories belong can be broken down ilyto a series of nested binary contrasts.

The tvw views of myth seem very contrary. Levi-Strauss I s view of myth
draws attention to the common functions of human thought in organizing
human experience ; Leach I s view dra',vs attention to the social (or political )
significance of cultural activities (such as myt.h--telling). But both
views dray; a -ttenLion 'co a point I wan'c to emphasize here. In l',fghanistan
sectarian distinct.ions may be more re s istent to change than ethnic dis·­
tinct ions i they. may be more deeply erru':1'eddcd in the l'.fghan psyche - and
hence may be for some purposes more "important" - than t:he ethnic distinc­
tions we were convened to discuss here.

The myths I shall tell here may be called "hero" myths. They focus on the
acts and p e rsonal qualiti e s of the horoes of the three sect groups. Afghilli
sectarian hero myt~hs resemble some of the myths tha'c are told among the
various Kachin comrnunities in highland Burma 'chat Leach studied. They
describe events a.nd r e l ationships in the lives of past heroes, and by im­
plication -they represent the different claims of contempor2ry groups of
people upo n highly desired resources. Amons the Kachin a commonly told
myth was phrased differently by different rival groups so as to justify
their vu.rious claims to superior authori t:y. "Hi thout seriously altering
the structure of the mythological story, each of the given clans named ­
as weill as several others .- can put fo:n-,r a rd a cO.se to be regarde d as the
senior group" (Leach 1954: 271 i cf. ., pp. 264-278). Th e hero myths told by
the rival sect groups in Afgh a nistan are similar , except that the heroes
are not ancestors - or at least they are not venerated because they are
lineage heads. Th ey are venerated as prominen'c figures in Islam.

The heroes of a Huslim sect are the persons who once championed its
causes or otherwise stood for the sect in a major sectarian quarrel. The
three sec-t:s in Afghanistan hu.ve historically diff e red over which persons
had the right to lead the Huslirn community. lUmost

36
ach
immediately after the Prophet's death a dispute arose over who should
3 are I
succeed him as the leader of the t·1usl im community. The disp ute has con­
lde
tinued in diff erent guises until the present time. Each of the three
For
sects in Afghanistan holds a dogmatic position on the historic quarrels
any
over Islamic leadership. Sunni Muslims believe the rightful succe ssors
Jth
of Huhammad ,,rere the first four Caliphs - ]I,bu Bakr r Omar, Osman and Ali ­
teller's
known as the "four fri ends " (of l.'luharmnad). 'They are regarded by the
and
Sunnis with specia l reve.-cenee because of their close a ssociation with the
(ibid.,
Prophet. But Sunnis do not regard the Caliphs that followed Ali with much
d:
esteem because of their personal and moral failures and because they
) binary
consistently flout ed the advice offered by the religious scholars of their
thought.
times . Those Caliphs "rere posthuJTIously stripped by Ottoman theologians of
primary
the religious titles they enjoyed in life.
,,!hich
.rily of
Shii tes, in contras t to the SUllnis 1 believe that from the beginning ·the
. opera.tes
only rightful successor of the Prophet was Ali and consider the other
'c:s·ta·tions
three no bet~ter th an usurpers. Ali's supe rior right to succeed the Prophet
In myth
wac; based on his close fami. ly rel a·tionship to 'ch.e Prophet; he was Muham­
.t
mad's cousin and, by hi s ma rriage to Fatima, his son-in-law . So Shiites em­
these
phasize aut.horitY deriving Leom fami ly ties \"iith l'luha.mr~,acL 'I'hey especially
ltrasts .
venerate the "five persons, II the pa12;L.!:.~.~: ("five fingers") f i_' s=...;; Huh cu1h'Tlad ,
Fatima r Ali and ·their tvo sons r Hasan and Hosayn, Ali's sUCC:CSSOJ:S, they
7th
say , should have been the senior mal e s in Ali's family line. These they
Lng
call ".!mc:m " (inst.ead of "Ca liph"). But the Shiite s hilve div.i .c1ed among
Litical)
thernselyes on "'ho the r eed . Imar.,s ';Icre .- that is, on ""hi ch lin e age of Alid's
th
was primary and therefo r e authoritative . The t vo Shiite sects of Afghani~can
1istan
agree on the line of Imams after lUi d ovrn to the seventh genej,-atio n. 'rhey
:3.is··­
differ over who should have been r ecog nize d as the seventh Imam. The sec-t
and 'knol"{n as Imamis (or "Twclvers") follm·1 on8 son of the six·th Imam and a line
istinc­
of his descendants that ended wi t.~l the 'c,·v elfth 1m.CUll ivho disappeare d ; they
now m·,rai·t his return at the end of this age. The other sect , t:he Ismo. ilis f
follow anothe r son of the sixth Imam and a line of his descendants known
on the
to us as the " Aga Khans."
Afghan
* * .,;
the
hey
These differen,::;es over the rightfu l snccessor of r,juhammad. are evident in
by im­
the stories the Sunnis and Imami Shiites tell about the early Muslin heroes.
S of
The Sunni and Imami Shiite hero myths that £0110'.",) wer e told to me by a
told Shiite mullah. He seems to h ave bC8n a fairly good source of information
tity
about 'che belie fs of both. groups. Pe grew u.p as a Shi ite, reading Shiite
:ering
books as a young man to his family and friends; he lived for several years
ilTled ­ in a dormi-tory v1ith Sunni boys whil e in grammar school, where he read
lS the
Sunni books. Today he says he does not believe in any religion.
:old by
leroes The sectarian myths I heard from him are of two types: those that describe
, are
heroes as extraordinarily good or strong (these are the stories one tells
about tIle hero es of one's own sect), and those that describe hero es as
extraordinarily wicked or nasty (t.hese are the stories one tells about the
heroe s of a rival sect). I shall describe a.nd illus·tratethese two types
L. The
of hero myths c.n d th en discuss what this CO!lt:rc:s t "says" about t:he
"evolving interfaces" of socio --poli tical groups in Afghanistan.

37

"Nice" H13ro Hvths. The Sunnis of course tell stories about the superior
piety and power of the "four friends" (Abu Bakr, Ornar, Osman, Ali). Here
is an example, translated for me from a religious book, Gulshan-i-Arifin,
by Mowlana Sultan Aref, that I purchased in Kabul.

Hazrat-i-Ornar was full of courage. All real believers have

courage. When Omar conquered Egypt he appointed a just gover­
nor tNere. The next year there was a drought and the Nile
River was almost dry. The people went to the governor and
told him that every year {when there was a drought] they threw
a beautiful maiden "\.0 the god of the Nile. The governor wrote
to Omar and Omar wrote to the Nile River saying, "Rise, by the
name of God," and the river rose.

A similar story is told about Abu Bakr, presumably to legitimize Abu Bakr's
claim to the Caliphate.

vfuen Huhanunad went up into the seventh heaven [my Shiite friend
said] God said to him "Asalaam Alaykum " and Muharrunad said,
"hTa'leykom 'Asalum;" Gabriel was with him and he was very
afraid but Muhammad said, "Don't be afraid, because I am going
along with you." There was a white curtain there and a tray of
palaw [the Afghan I s most popular food]. Then a hand carne from
under the curtain and said, "I will eat with you," and the hand
ate with Huharrunad. Sunnis believe the voice that spoke to him
was much like that of Abu Bakr's. It was supposed to be God's
voice but it ,vas like Abu Bakr' s. When Huhammad came back to
him and Abu Bak.r said "Salaam Alaykum" to him, 11uharrunad recog­
nized that it Vias the voice of Abu Bakr. He is like the tongue
of God, they say, because since God has no mouth He talks for
God, like Moses did.

The Shiites in a similar ",ay venerate Ali. The Sunnis do not dispute this,
of course, as they include Ali among the "four friends." It seems under­
standable that the stories about Ali's powers abound in Afghanistan, as
there is no one to dispute them. It is said that Ali had superhuman power,
like Omar, that he made the Band-i-Amir lakes, that he killed "dragons"
in Bamian, and so on. One story about Ali's powers you might not have
heard is this:

The s.ister of Hazrat-i-Ornar was given to Ali one night [my

friend once told me]. An angel came to her and told her she

was to marry him and a power not her own brought her to him

that night. Ali himself said the nek~ [Islamic marriage

rites] and the four corners of the room acted as the four

witnesses [required by Islamic law for legal marriage]. And

that night Ali had sexual relations with her seven times.

Next morning Orna r saw that there were seven washcloths there

lrequired for cleansing after sexual relations, according to

Islamic law]. And he saw his sister reading a Quran. He said
to her, "\'lliy have you become a Muslim?"

Also, Ali had, Shiites insist, the preeminent right to be the first Caliph

38
 According to Shiites, when Muhammad ascended into heaven he found evidence
ior that Ali - not Abu Bakr - belonge d there.
Here
fin, When Muhammad was ente ring this place (heaven) he was stopped
by a lion and Gabriel told Jlluham_'Dad he would have to give him
something or he wouldn't let him go. So MuhaIiunad gave him his
ring vlhich had a ruby stone on it and the lion let him go in.
This is why it's good to pray Hith a ruby ring on your right
hand. The next morning \"hen Muhanunad came to earth he vlent to
the mosque to pray and on the \'lay Ali met him and asked, jokingly,
"hThe re haVe"" you been?" and h e gave him his ruby ring. This is
why Ali is called "The Lion of God." Jl1uhammad said, "Now I know
you are the lion of God."

Sunnis and Shiites alike venerate Ali. But as you can see, they differ on
Bakr's whether Ali had any place in the he avenly reCllms when Huharrrrnad visited
there, for the Sunnis believe it was Abu Bakr that MuhClTIunad encountered
there. By impli catio n, of course, the >cwo vers ions of this s tory indi cate
that they disagree on who had the righ t to be the first Caliph. Shii tes
and Sunnis differ on some other matters concerning rightful success ion, for
as I said earlier, the Shiites venerate some hero es the SW1I1is do not
recognize. The Shiites especially cherish the memory of Hosayn, "'ho along
with 69 other persons was cruelly murdered by Sunni troops in the early
days of I slam . The Shiite mullah told me th e following ahout Hosayn.
The story resembles a Christian -theme: 1

There is a c hapter in one of my book s [which he had inherited

from an uncle] about Hosayn's acceptance of the death of his
relati v es before 1::he world ,-Jas created. God did many -things
before the beginning of the world. He wanted to t es t people's
faith and also to see- who would bear the most suffer ing.
There vIas a glass of poison in 'dhich all th e troubles o f the
world were mixed. Adam took somc, then other s took some i but
~e this, they did not take i t all and the worst of it was in the bottom.
mder­ Then it came Hosayn's turn. So h e sa id, "I \'7ill take it all.
as 1 will give all so tha-t Huharrullad's followers ,vi11 be forgiven."
1 pO'.ver,
And God S':lOre that on Doomsday He ",ou1d forgive so many of
~ns "
Muhammad's fol1ovlers that Fatima ItTou1d be happy and say, "0
lye God, this is worth the price of my son's blood. " God will
make her, Hosayn and the others [of l'luhammad' s family] happy
by forgiving peopl e because they h ave given th e ir blood for
the sins of Huhanunad' s follo",ers >co win forgiveness for the m.

1
Compare the follol'ling vers es from the h'ew Testamen t:
"Father, ... remove this cup frornmei nevertheless ... " (Hark 14:36)_
"Since the children share i n the flesh and blood, he J esus himself
likewise partook of the same nature, that through death h e might destroy
him who has the power of dea-th, that is, the devil, and deliver all
those ",ho through fear of death wer e subject to lifelong bondage."
(Hebrew s 2:14,15).
"hlhile we were still helpless, at the right time Christ di ed for the un­
godly ... - God shows hi s love for u s in that. while we i-,ere still sinners
Caliph. Christ died for us ... ther efore , we are nm'l justified by his blo00 _ i
(Roman s 5: 6,8,9).
"Re himself bore our sins in his body on the tree, that we die to sin an (1
live to righteousness. By his wounds you have been healed."

-~- . 39
--­
(I Peter 2: 24),
"Nasty" Hero l-lyths. All these stories emphasize the unique and special
powers of se ctarian heroes. But tra di tional Afghans also tell stories
defaming the heroes of rival sec t groups. Sunnis seldom defame th e
early and illustri~)Us her08s of the Shiites but they do defame less
prominent and subsequent Shiite heroes, ,,,hom we sometimes call their
"saint_s," mo s tly to indic a t e that they are impos·ters. I omit here the
Sunni pejorative stories about Shiite saints in order to avoid the ex­
tended discussion of JI:Ius lim sai nts that that ,'lOuld require. But with
regard to the gTeat ear ly heroes of I slam already mentioned above, the
Shiites have l ong def amed all the gr eat heroes of Sunnism 8xcept Ali.
As a matt_er of dogma l~he Shiites heap sco rn upon t.he first three C",l iphs ,
Abu Bakr, Omar and Asman r for Shiite dogma insist_s th at ]>.l i and his family
alone have the right to lead the Buslim community. Shiite ridicul e of the
ear ly Caliphs of course angers Sunnis inte nsely. A Sunni'rulcr of Afgh an­
istan, Amir Abdul Rahman, on ce banned Shiite books for this reason (Kakar
1974: 303--309) _ Here is an ey;amp le of hm'l the Shiites defame Osman, as
told by the Shiite mullah:

Shias believe Osman killed one of his wives who vTaS a dangh-ter of
Huhinmnad. He beat he r wi-th a heavy stick from a camel's chair.
She had gone thr ee times to Ivluha lTcmad and told him th a -t Osman
was beating her cwd she was aIJ:-aid he might kill her. But Jl1uham­
mad wanted to show the right of a husband over th e wife so he
sen·t h er back to her husband. l\fter that lone of HuhamrClad' s
e nemi es - he wa s a cousi n of Osman who in one of the battle s
broke -the tooth of i"11lh arnm3.d - thi:3 man co_i.ne to Osman's house.
Gabriel came to Muh a.mmad and told him: "The :man HhCl broke your
too-th is in Osman's hous e. Nm·[ send Ali \'ii th this s,mrd." This
sword, according to the Sunnis , came from God to Muh 2mull ad in a.
box b ut o_cco:cd.ing t o t:h e Shias it was brought only for special
purposes to the family of i'iuharmna d by the ange l Gab riel . Muham­
ma d sent Ali hut_ [when he got the:cel Osrr:::<n 'cold him there was
no one there, He had hidden h is cousin under a camel sadd l e .
vJhen Ali left Osman sent his cOl)::; in out _ on a camel and gave him
water, Btlt on the ':1cl'I the came l became sick and couldn ' t go.
It j'ust le id dOVlrl. So the mem beq a n to walK. Th en his s hoe
st:caps broke,". A bird caIne and took his f ood from him and another
bird put his beak through the ski n th at held the vlat er. So
finally h e wa~3 crchvling and about to die. Gabriel Ciune
and told ]'h.lhc:: mmad to send 7->.li to a certain place t .O tind that
man and Ali found him and killed him th e~-e a nd brought hi3 head
back with him. But when Osman he ard it he ac cus ed his wife of
telling Ali ,'There the man VlaS - he didn't believe in Huh arrunc::.d ' s
spiritual powers. So he beat heJ~ until she d i ed .

* * *
Th e contr as t betw ee n t he hw traditions of myth telling can be charted
to sho\o, the catego rical differencesilT'plied in the bw traditions. The
myth s differ as to vlhether they a r e appro ving ("nice h e ro e s") or disap­
proving ("nasty hero es ") and as to whether the speaker is Sunni or Shiite.
(Figure 1).

40

 Figure 1

AFGHAN' HERO HYTHS

OWn Heroes Outsiders' Heroes

Sunni Imami Re: Imami Re: Sunni

Ihs ,
heroes heroes
mily Omar: Ali:
: th e Nile Virility
:ghan­
Osman:
tkar
Abu Bakr: Ali: wife killer
lS
in heaven in heaven

Hosayn:
Atonement

~=' ----- ------- -- ---~ ------- -.-------

-------------.--

Observe that. the contrasts are binary. Levi-Strauss, I mentioned, sees

in myth evidence of the dualistic workings of the hwllan mind. Structuralists
seem to find it almost everywhere. Consider the similarity betvleen the
binary structural contrasts implied in these Afghan myths and the con­
trast that Haybury-Lewis described in the structure of Shavante society.
Maybury-LeHis argued that,". '. . Akwe-Shavante s:)ciety can best be understood
in terms of the dichotomy betv,'een waniwiha [insiders] and wasi' re 'wa [out­
siders]" (1967:299). In this sense the structure of Afghan society may
correspond with t/jaybury-Levlis' s concept of a "dual organization\; as "an
ideal type corresponding to a theoretical society in which every aspect
of the life of its members is ol-deredaccording to a single antithetical
formula" (1967: 298). He puts it that "the best explanatory mod e ls for
Shavante society are all dyadic ones If (1967: 299). The structure of the
Afghan worlclview, as manifest in these h e ro myths, ,-lOuld seem to b e as
plainly dualistic as t.hat of the Shavante. This would, I suppos e, please
Levi-Strauss: once again the pervasive influence of the binary form of
thought is revealed!

NoW consider the social significance of the contrasts that these myths
reveal. They exemplify th e rivalries between the sect groups in Afghanistan.
r think these rivalries ought to receive closer attention when we try to
assess traditional socio-political relations in Afghanistan society.
:d
These sectarian contrasts are contemporary. The old wars of the past,
'he
with their sectarian overtones, persist in tales of sectarian heroes.
lP­
Many Afghan heroes are religious figures. The moral domain of social af­
liib:; . fairs is imbued with religious ideals e~pressed not only in religious
dogma and ritual but also personified by e xemplary, even divine, heroes.
The traditional heroes of the Afghan people are, on the whole, religious
personalities - Ali, Hosayn, Abu Bakr. There are, to b e sure, other
Afghan heroes like Ahmad Shah Baba or Khushal Khan Khattak. But the heroes
that, it seems to me, commonly inspire the admiration and strong feeling
of fuost Afghans are religicius.

41
And if this is so, then how shall we understand traditional Afghanistan
society? What issues do the Afghans hold most dear if not religious
ones? And if so, then what differences do they hold most tenaciously?
The underlying bonds of trust and respect - how do they fall? I think,
again, they fall along sectarian lines. The Afghan people marry within
the sect, they socialize within the sect, they buy and sellon a regular
basis within the sect (although less so now in the cities, cf., Canfield
1973). I wonder if as much could be said for ~~hnic distinctions. People
tend to ma~ry within their ethnic group, it is true, but they do not marry
or socialize with, or buy and sellon a regular basis with, members of
their own ethnic group unless they are members of the same sect.

I . do not mean to insist that sectarian distinctions are ah/ays for all pur­
poses more significant in Afghan social affairs than ethnic differences ­
though I personally suspect they often are - but I do mean to say that no
study of Afghan traditional socio-political alignments can ignore the
sectarian differences. We must not assume that sectarian alignments are
somehow irrelevant to politics, including ethnic politics, simply because
the Afghans are no longer fighting sectarian wars, nor should we assume
that sectarian alignments are now politically benign. Traditional cultural
differences still impinge on social life, the old sectarian contrasts
as well as the ethnic ones. The centuries-old quarrels between Muslims
still blemish the realm of social trust. The stories that Afghans tell
about their heroes reveal that those old blemishes persist, still there
to derail · social relations between people. Students of Afghan society
must be alert to the ways that this tradition a.ffects social relations,
including etlmic ones, today. Afghanistan may be changing and with the
change the old alignments may be being transformed. But the old
sectarian alignments are not dead.

42
 ~----

n
REFERENCES CITED

"
Abdur Rahman Khan. 1900. Th e Life of Abdur Rahman Khan, Amir of
n Afghanistan . Sultan r1ahome d ' Khan , ·--ed.~- London: john Murray
ar

d Adamec, Ludwig. 1967. !f~0anistan, 1900-1923. Berkeley: Univeristy

:ople of California Press

larry
Akhramovich, R.T. 1965. 2.~t.~~!:"'~..9_~_l-!'_ist.0ry of Afgha_~_i s t:_an a_~ter the
Second lvorld War. r.loscmv: Neaka Publishing House

pur­ Allworth, E. 1973. ~~ __Nati_~nali ty--.9~.~~~_~~ in Soviet Central Asia.

~s - New York: Praeger
: no
Anderson, Jon W. 1975. Tribe and Community among Ghilzai Pashtun.
ire Anthropos, 70: 575-601
ms e
Ie 1976. Structure and Power in Pakhtun Tribes. Conference on , Rural
.tural Life in Afghanistan, University of Nebraska at Omaha, [forthcoming
in R. Newell, ed., ~ur~~ife in A~gh ~~ist~~~~0~ Pros~ects for
Development] .

~e
Bacon, Elizabeth. 1951. An Inquiry into the History of the Hazara Mongols
of Afghanistan. Southwestern Journal of Ant~~opology, 7: 230-247.
; ,
1958. Obok: A_? tudy of Social Structure in Eurasia. New York:
Wenner-Gren Foundation.

Barth, Frederik. 1956. Eco~ogic Relationships between Ethnic Groups In

Swat, North Pakistan. !IDeric ~!l Anthropologist t 58: 1078-89.

1959. politic~l Leadership ~~ Swat Pathans. London: Athlone Press.

1960. The System of Social Stratification in Swat, North Pakistan. In

~ects of Caste, E.R. Leach, ed. Cambridge: Cambridge University Press.

1969. Pathan Identity and Its I1aintainance, In_~th~i-~_Gro~p..§~p.d

Boundari~, F. Barth, ed. Oslo: Universitetsforlaget.

Bellew, Henry W. 1891. ~~_In3 ~¥--'!'-.~~~_~0~ Ethnograph¥ of Afghanistan.

Woking: Oriental University Institute.

Burnes, Alexander, R. Leech, P.B. Lord & J. Wood. 1839. Reports~

Papers, Political, Geographical and Commercial on SciJ:Cle, Afghanistan
'aJ;'d-=]I;~:racent - ' C01!-ntries_. ' C-alcutta:--Be ngai-~1ni taryPiess . - -- -­

(C.A.M.). 1894. Central Asian Materials, I~ (collected by the editor) .

.J;.Inp. ~ r:~a~_ Cl:!2.9.. l\_s.iat:.~~_Q~a~_t_~~~y_~~..YJ:.~~_?:.nd Or~ental and Colonial Record.
8: 407-420.

Canfield, Robert. 1973a. F'?c.t:i.gl)_c!}1c'L~9})v ~ rs:i.9.l1.. _i.!La. _121'Q_:r.?_L_SQ.c;;j..et...Y:

Re.1:.~.9~<2..u_!3__Alj gmnel:,1 ts in the Hin<llL..Kush. Anth. Paper No. 50 .. Ann
Arbor: University of Michigan Museum of Anthropology.
1973b. The Ecology of Rural Ethnic Groups and the
Spatial Dimensions of Power. l\m~.EJ..~ _ Al}_!=:h£2P..9) g.9i.~t, 75:: 1511-1528.

43
Cohen, Abner. 1969. Custom and Politics in Urban Africa. Berkeley:
University of California Press.

Coon, Carleton. 1958. f~_~_~an?:, ___Th~_,~torLof_j:_he Midg._~,~_~ast. New

York: Harper and Row.

Dupree, Louis. 1956. The Changing Character of South-Central Afghanis­

tan Villages. ~~~_ Organization, 14:26-29.

1964. Tribal Traditions and Modern Nationhood. Asia, 1: 1-12.

1973. A~9hanistan. Princeton: Princeton University Press.

1975. Settlement and Migration Patterns in Afghanistan: a Tentative

Statement. l:192-_~~n A~,~_Studies, 9:.397-413.

Elphinstone, Hountstuart. 1815. An AccOl.Lnt o ;L:t:he Kin9~!Il_~ Cc:ubul.

London: J'ohn Murray.

Entienne, Gilbert. 1972. ,~~~:f._9..~~.!!,~,.~_~~n~1d les l\l.,eas _~~"l-,a C,?operation.

Paris: Presses Universitaires de France.

Farhadi, A.G. Rawan. 1970. Languages. Kabul Times ,Annual. Kabul:

Gove~nment Printing House ,

Ferdinand, Klaus. 1962. Nomad Expansion and Commerce in Afghanistan.

Folk, 4: 123-159.

1964. Ethnographical Notes on Chahar l',imaq, Hazara and Moghol.

Acta Oreintalia, 28: 175-204.

1969. Nomadism in Afghanistan, In Y,iehwix:!sc}:,~,ft un~!..irtenl~l}} t:~~,

L. Foldes, ed. Budapest: Akademia Kaiao.

Fletcher, Arnold. 1966. _,~.9b~~~~tan:_Highway of Conquest. Ithaca:

Cornell University Press.

Franck, Peter, . 1955. Technical Assistance through the united States­

U.N. Mission in Afghanistan 1950-1953, In Hands Acro~~J~'~O}2:tie_rs I
Howard Teaf, ed. Ithaca.: Cornell University Press.

Franz, Erhard. 1972. Zur Geg e nwaertigen Verbreitung und Grupperiung

der Turkmenen in Afghanistan. Baessler"Archiv, 20: 191-236.

Glatzer, Berndt. 1977. Nomaden von Gharjistan. Wiesbaden: Franz Steiner

Verlag.

Gregorian, Vartan. 1969. T~~__ ~!~er9E;,!lCe of Mo~ern AfghanistcUl. Palo

Alto: Stan~ord University Press.

Groetzbach, Erwin. 1972. ~~_~!1.J~g_E!9.9}:~ap.h:i§~£le WClI!c=l~.1_ in _Nordost­

,~!.9hClI!J:~ !,aIl--'" ei t __ g e m 19. ;a0rh~p_dert. Afghanische Studi e n nr. 4.
Meisenheim am Glan: Anton Hain Verlag.

Gulick, John. 1976. Jh,f?-.!J.tClg.J:.~_J~asj:-=-_l'l.!l_l!n:tJJroIlo),()_9:.i~_Cl._LPe zsp,ectiy~.

Pacific Palisades: Goodyear Publishing Co.
44
 ~y : Janata, Alfred. 1962/63. Die Bevoelkerung von Ghor. Archiv fuer
Voelkerkunde, 17/18.

~w
1971. On the Origin of the Firuzkohi in Western Afghanistan.
M2f1_:!:.Y fuer Voelkerk~nd_~, 25: 57-65.

~ anis- Jarring, Gunnar, 1939. On the Distribution of Turk Tribes in Afghanistan.

Lunds Universitets Arsskrift, Bd. 35, no. 4.

Kakar, M. Hasan. 1971. _h..:f:9b.i:l}2i_stan...:._A Study in Internal Pplit,ica1

pevelopments, 1880-1896. Lahore: Panjab Educational Press.

1974. Afghanistan During the Reign of Amir Abd AI-Rahman Y~an

:1tative 1880-1901. Ph:D. Thesis. University of London.

Khanikoff, H. 1845. ~2~f1-.i:l'!:.9_:....1J_!?__ ~t..J;"_i:lncL~~ople trans., C. de Bode.

DUL London: James Madden.

Kushkaki, B. 1923 . _Rahmi:l..!1_,,=::i -Qa:t:.i:l9.b2!LWS!.J2.Cl99-J~hshal) [Guide to Qataghan

ation. and Badakhshan]. Kabul: Ministry of Defense Press.

Leach, Edmund R. 1954. J'oli_tic.a} _.~¥s!:~!."rls of Highla.nd _Bu~El:' Boston:

1: Beacon Press.

Levi-Strauss, Claude. 1963. Totemism. Boston: Beacon Press.

MCl.cGregor, C .M., ed. 1871. ~~nt.~3':! .J:l.si~-,--.!'_ar~ __~I :._~_~_<2.n~!ib_'=!!:_~.9ll--'£..C?~_~rC!.§.

tJ1_E?__B e t_t:~~._ !'CJ:'!.9~~J_~_qg~.9.t __th~_._'!'2Eogri:lJ2QyL Ethnol~gy , Reso ~rc_~'::>_.~_J1_d
1. Histo.rr Q.Llifgl1anista,.n. Calcutta: Government Printing House.

Maybury-Le\vis, David. 1967.

~~"~-=§h~D!e Society. Oxford: Oxford

University Press.

Miran, Mohammed Alam. 1977.

Sociolinguistic Factors in Afghanistan.
a: !\f.9E.9.!!istan Journal, 4: 122-127.

Newe ll, Richard S. 1972. ~o~.Ltj~'§_9_L~fg}lar:,-~.~t9.l2. Ithaca: Cornell

tes­ Unive rsity Press.
ITs,
Poullada, Leon. 1973. R~fo_rm_~l)sL.Reqe~lLQD__.iJJ.._.h:f:.9.h~[lis..:tan J,21~-1~.
Ithaca: Cornell University Press.
ung
Rosman, Abraham & Paula Rubel. 1976. Nomad-Sedentary Interethnic Relations
in Iran and Afghanistan. In.:t~E!l._att9.n?lJourn?J.._..?.l 11iddl~ ~ast
Steiner Studies, 7: 545-570.

alo
Simmel, Georg. 1950. The Stranger, In '!'b.§.._}~,-<?cioJ,9.9Y_Q.i Ge~§Jmmel,
K. Wolff, ed. Glencoe: Free Press •

. 4.

45

.... -
 Shahrani, M.N. 1974. The Retention of Pastoralism among the Kirghiz of
the Afghan Pamirs. Presented at the 73rd Annual Mee ting of the Ameri­
can Anthropological Assn., Me}:ico City. [to appear in J:!jmaJ:_~ll
A~1:1.2rqEologZ. : __.1'p_e__I12do-1'?:betan Interface, J.F. Fisher, ed. The
Hague: Mouton].

1976a. Kirghiz Pastoralists of the Afghan Pamirs: An Ecological and

Ethnographic Overvievl. FolK, 18: 129-143.

1976b. Kir0hiz Pastoral Nomads of the Afghan Pamirs: A Study in

Ecological and Intra-Cultural Adaptation. Ph.D. Dissertation,

Unive rsity of Washington, Seattle.

forthcoming. ~dapt_a,:Lion_ t~~l~§~~_. Fr<..?}!J:t~_§....= . _,!,_b.~___Ki:f9biz_ and \vakhj..

QLb.fghcLIl isJ:...?-D. Seattle: Uni versi ty of Washington Press.

Slobin, Nark f, Greta Slobin. 1969. Ethnography of Afghanistan: A Rus s ian

Study by M.G. Aslanov, E.G. Graffenb e rg, N.A. Kisliakov, K.L.
&G.P. Vasilyeva, In _Afgh ~~is~E~Some-p e\~22rQE. che § , G. Grassmuck,
W. Adcmec & Francis Inlin, eds . Ann Arbor: Center for Ne ar Ea s tern
and North African Studies.

Steul, willi. 1973. Eigentumsprobleme innerhalb Pasch t:u_nj..~ che_:r: . . GeI!le..:tn­

~ch_a:tt:_~I',--. j..}LPa~t i. cVA f9han;:Lst ~m.
Heidelb e rg: Suedasia n Insti tut (m

Strand, Richard F. 1973. Note s on the Nuristani and Da rdic Language s.

Jo:grll.a l of ~. A merican Ori e ntal . So~j~, 93.3: 297-305.
. . ,

1974a. Principles of Kinshi p Organization among the Kom Nuristani,

In Cultures of the Hindukush: Selected Pa.l:?ers_J),om t.bJL Hi nd1,1- ~1t~h

Cultm·ai- Conf-er~nc e-- H e id at }1;;esg;;di9 ·7 0 ,-"K~~rl Jettmar, ed., in

;;;-ilab~;-at i~~--::Yi t:h Lenna~t- Ed e lbe rg. Be i tri:i.ge zur Su'dasienforschung,

S~~asien-In s titut, Universit~t Heidelberg, 1: 51-56. Weisbaden:

Franz Steiner Verlag.

1974b. A Note on Rank, Political Leadership, and Gove rnme nt among

Pre-Islamic Kom, In Culturg .f?_ Qf ._t.b~_J:l..:i".D..@ .tu?.h (s e e above ), 57-63.

1975. Th e Changing He rding Economy of the Kom Nuri s tani.

Journal) 2 (4): 123-134.

Swidler, Warren ~v. 1973. Adaptive Processes Regulating Nomad-S e dentary

\ Int e raction in the Niddle Eas t, In !'b..e Des_C:=E1::_...9..!l_C! . .th e Sown, C . . Ne

'.
'. ed. Berkele y: Institute for International Studies.

apper, Nancy.

Afghanista n.

36: 55-79.

Wood, J. 1941. A. P~r§c?!2.9_J, _ Na g2. tive of a Journey'~~_ the.~o~rceyt_the

Riv.~.! Q~~ . London: John Nurray.

46